moebiusband/MD-Bench
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

added static analysis and likwid files

Browse Source
This commit is contained in:
JanLJL 2023-02-09 17:33:22 +01:00
parent 3b076cdb49
commit cb5598bc91
11 changed files with 11709 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

88
likwid-outputs/csx-lammps-dp-mem_dp-stub.out Normal file
View File

@ -0,0 +1,88 @@
--------------------------------------------------------------------------------
CPU name: Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz
CPU type: Intel Cascadelake SP processor
CPU clock: 2.49 GHz
--------------------------------------------------------------------------------
Initializing parameters...
Initializing atoms...
Creating atoms...
Pattern: seq
Number of timesteps: 200
Number of atoms: 256
Number of neighbors per atom: 1024
Number of times to replicate neighbor lists: 1
Estimated total data volume (kB): 1062.9120
Estimated atom data volume (kB): 6.1440
Estimated neighborlist data volume (kB): 1050.6240
Initializing neighbor lists...
Creating neighbor lists...
Computing forces...
Total time: 0.2735, Mega atom updates/s: 0.1872
Cycles per atom: 10682.8568, Cycles per neighbor: 10.4325
Statistics:
Vector width: 8, Processor frequency: 2.0000 GHz
Average neighbors per atom: 1018.9055
Average SIMD iterations per atom: 127.3632
Total number of computed pair interactions: 52428800
Total number of SIMD iterations: 6553600
Useful read data volume for force computation: 1.47GB
Cycles/SIMD iteration: 83.4598
--------------------------------------------------------------------------------
Region force, Group 1: MEM_DP
+-------------------+------------+
| Region Info | HWThread 0 |
+-------------------+------------+
| RDTSC Runtime [s] | 0.110776 |
| call count | 200 |
+-------------------+------------+
+------------------------------------------+---------+------------+
| Event | Counter | HWThread 0 |
+------------------------------------------+---------+------------+
| INSTR_RETIRED_ANY | FIXC0 | 267036300 |
| CPU_CLK_UNHALTED_CORE | FIXC1 | 219034500 |
| CPU_CLK_UNHALTED_REF | FIXC2 | 273793400 |
| PWR_PKG_ENERGY | PWR0 | 10.9296 |
| PWR_DRAM_ENERGY | PWR3 | 0 |
| FP_ARITH_INST_RETIRED_128B_PACKED_DOUBLE | PMC0 | 0 |
| FP_ARITH_INST_RETIRED_SCALAR_DOUBLE | PMC1 | 159400 |
| FP_ARITH_INST_RETIRED_256B_PACKED_DOUBLE | PMC2 | 0 |
| FP_ARITH_INST_RETIRED_512B_PACKED_DOUBLE | PMC3 | 197068800 |
| CAS_COUNT_RD | MBOX0C0 | 8643 |
| CAS_COUNT_WR | MBOX0C1 | 1367 |
| CAS_COUNT_RD | MBOX1C0 | 9124 |
| CAS_COUNT_WR | MBOX1C1 | 1354 |
| CAS_COUNT_RD | MBOX2C0 | 9138 |
| CAS_COUNT_WR | MBOX2C1 | 1356 |
| CAS_COUNT_RD | MBOX3C0 | 5586 |
| CAS_COUNT_WR | MBOX3C1 | 1297 |
| CAS_COUNT_RD | MBOX4C0 | 5328 |
| CAS_COUNT_WR | MBOX4C1 | 1269 |
| CAS_COUNT_RD | MBOX5C0 | 5280 |
| CAS_COUNT_WR | MBOX5C1 | 1295 |
+------------------------------------------+---------+------------+
+-----------------------------------+------------+
| Metric | HWThread 0 |
+-----------------------------------+------------+
| Runtime (RDTSC) [s] | 0.1108 |
| Runtime unhalted [s] | 0.0878 |
| Clock [MHz] | 1995.2564 |
| CPI | 0.8202 |
| Energy [J] | 10.9296 |
| Power [W] | 98.6643 |
| Energy DRAM [J] | 0 |
| Power DRAM [W] | 0 |
| DP [MFLOP/s] | 14233.3287 |
| AVX DP [MFLOP/s] | 14231.8898 |
| Packed [MUOPS/s] | 1778.9862 |
| Scalar [MUOPS/s] | 1.4389 |
| Memory read bandwidth [MBytes/s] | 24.9001 |
| Memory read data volume [GBytes] | 0.0028 |
| Memory write bandwidth [MBytes/s] | 4.5861 |
| Memory write data volume [GBytes] | 0.0005 |
| Memory bandwidth [MBytes/s] | 29.4863 |
| Memory data volume [GBytes] | 0.0033 |
| Operational intensity | 482.7104 |
+-----------------------------------+------------+

168
likwid-outputs/csx-lammps-dp-mem_dp.out Normal file
View File

@ -0,0 +1,168 @@
--------------------------------------------------------------------------------
CPU name: Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz
CPU type: Intel Cascadelake SP processor
CPU clock: 2.49 GHz
--------------------------------------------------------------------------------
Parameters:
Force field: lj
Kernel: plain-C
Data layout: AoS
Floating-point precision: double
Unit cells (nx, ny, nz): 32, 32, 32
Domain box sizes (x, y, z): 5.374708e+01, 5.374708e+01, 5.374708e+01
Periodic (x, y, z): 1, 1, 1
Lattice size: 1.679596e+00
Epsilon: 1.000000e+00
Sigma: 1.000000e+00
Spring constant: 1.000000e+00
Damping constant: 1.000000e+00
Temperature: 1.440000e+00
RHO: 8.442000e-01
Mass: 1.000000e+00
Number of types: 4
Number of timesteps: 200
Report stats every (timesteps): 100
Reneighbor every (timesteps): 20
Prune every (timesteps): 1000
Output positions every (timesteps): 20
Output velocities every (timesteps): 5
Delta time (dt): 5.000000e-03
Cutoff radius: 2.500000e+00
Skin: 3.000000e-01
Half neighbor lists: 0
Processor frequency (GHz): 2.0000
----------------------------------------------------------------------------
step temp pressure
0 1.440000e+00 1.215639e+00
100 8.200895e-01 6.923143e-01
200 7.961495e-01 6.721043e-01
----------------------------------------------------------------------------
System: 131072 atoms 47265 ghost atoms, Steps: 200
TOTAL 11.50s FORCE 5.28s NEIGH 5.91s REST 0.31s
----------------------------------------------------------------------------
Performance: 2.28 million atom updates per second
Statistics:
Vector width: 8, Processor frequency: 2.0000 GHz
Average neighbors per atom: 76.0352
Average SIMD iterations per atom: 9.9181
Total number of computed pair interactions: 2003182862
Total number of SIMD iterations: 261297661
Useful read data volume for force computation: 57.46GB
Cycles/SIMD iteration: 40.4432
--------------------------------------------------------------------------------
Region force, Group 1: MEM_DP
+-------------------+------------+
| Region Info | HWThread 0 |
+-------------------+------------+
| RDTSC Runtime [s] | 5.115807 |
| call count | 201 |
+-------------------+------------+
+------------------------------------------+---------+-------------+
| Event | Counter | HWThread 0 |
+------------------------------------------+---------+-------------+
| INSTR_RETIRED_ANY | FIXC0 | 12592470000 |
| CPU_CLK_UNHALTED_CORE | FIXC1 | 10196910000 |
| CPU_CLK_UNHALTED_REF | FIXC2 | 12746120000 |
| PWR_PKG_ENERGY | PWR0 | 307.9429 |
| PWR_DRAM_ENERGY | PWR3 | 0 |
| FP_ARITH_INST_RETIRED_128B_PACKED_DOUBLE | PMC0 | 0 |
| FP_ARITH_INST_RETIRED_SCALAR_DOUBLE | PMC1 | 79042240 |
| FP_ARITH_INST_RETIRED_256B_PACKED_DOUBLE | PMC2 | 0 |
| FP_ARITH_INST_RETIRED_512B_PACKED_DOUBLE | PMC3 | 8076039000 |
| CAS_COUNT_RD | MBOX0C0 | 22734550 |
| CAS_COUNT_WR | MBOX0C1 | 1147714 |
| CAS_COUNT_RD | MBOX1C0 | 22755180 |
| CAS_COUNT_WR | MBOX1C1 | 1144415 |
| CAS_COUNT_RD | MBOX2C0 | 22762780 |
| CAS_COUNT_WR | MBOX2C1 | 1129051 |
| CAS_COUNT_RD | MBOX3C0 | 22905660 |
| CAS_COUNT_WR | MBOX3C1 | 1143324 |
| CAS_COUNT_RD | MBOX4C0 | 22914860 |
| CAS_COUNT_WR | MBOX4C1 | 1169116 |
| CAS_COUNT_RD | MBOX5C0 | 22890220 |
| CAS_COUNT_WR | MBOX5C1 | 1180739 |
+------------------------------------------+---------+-------------+
+-----------------------------------+------------+
| Metric | HWThread 0 |
+-----------------------------------+------------+
| Runtime (RDTSC) [s] | 5.1158 |
| Runtime unhalted [s] | 4.0885 |
| Clock [MHz] | 1995.2508 |
| CPI | 0.8098 |
| Energy [J] | 307.9429 |
| Power [W] | 60.1944 |
| Energy DRAM [J] | 0 |
| Power DRAM [W] | 0 |
| DP [MFLOP/s] | 12644.6041 |
| AVX DP [MFLOP/s] | 12629.1535 |
| Packed [MUOPS/s] | 1578.6442 |
| Scalar [MUOPS/s] | 15.4506 |
| Memory read bandwidth [MBytes/s] | 1713.4438 |
| Memory read data volume [GBytes] | 8.7656 |
| Memory write bandwidth [MBytes/s] | 86.5003 |
| Memory write data volume [GBytes] | 0.4425 |
| Memory bandwidth [MBytes/s] | 1799.9442 |
| Memory data volume [GBytes] | 9.2082 |
| Operational intensity | 7.0250 |
+-----------------------------------+------------+
Region reneighbour, Group 1: MEM_DP
+-------------------+------------+
| Region Info | HWThread 0 |
+-------------------+------------+
| RDTSC Runtime [s] | 5.897385 |
| call count | 10 |
+-------------------+------------+
+------------------------------------------+---------+-------------+
| Event | Counter | HWThread 0 |
+------------------------------------------+---------+-------------+
| INSTR_RETIRED_ANY | FIXC0 | 18212540000 |
| CPU_CLK_UNHALTED_CORE | FIXC1 | 11728500000 |
| CPU_CLK_UNHALTED_REF | FIXC2 | 14660630000 |
| PWR_PKG_ENERGY | PWR0 | 338.9000 |
| PWR_DRAM_ENERGY | PWR3 | 0 |
| FP_ARITH_INST_RETIRED_128B_PACKED_DOUBLE | PMC0 | 0 |
| FP_ARITH_INST_RETIRED_SCALAR_DOUBLE | PMC1 | 6240402000 |
| FP_ARITH_INST_RETIRED_256B_PACKED_DOUBLE | PMC2 | 0 |
| FP_ARITH_INST_RETIRED_512B_PACKED_DOUBLE | PMC3 | 983040 |
| CAS_COUNT_RD | MBOX0C0 | 2086787 |
| CAS_COUNT_WR | MBOX0C1 | 1115626 |
| CAS_COUNT_RD | MBOX1C0 | 2089964 |
| CAS_COUNT_WR | MBOX1C1 | 1117021 |
| CAS_COUNT_RD | MBOX2C0 | 2103832 |
| CAS_COUNT_WR | MBOX2C1 | 1117965 |
| CAS_COUNT_RD | MBOX3C0 | 2086930 |
| CAS_COUNT_WR | MBOX3C1 | 1102471 |
| CAS_COUNT_RD | MBOX4C0 | 2094688 |
| CAS_COUNT_WR | MBOX4C1 | 1103018 |
| CAS_COUNT_RD | MBOX5C0 | 2097438 |
| CAS_COUNT_WR | MBOX5C1 | 1102525 |
+------------------------------------------+---------+-------------+
+-----------------------------------+------------+
| Metric | HWThread 0 |
+-----------------------------------+------------+
| Runtime (RDTSC) [s] | 5.8974 |
| Runtime unhalted [s] | 4.7026 |
| Clock [MHz] | 1995.2473 |
| CPI | 0.6440 |
| Energy [J] | 338.9000 |
| Power [W] | 57.4661 |
| Energy DRAM [J] | 0 |
| Power DRAM [W] | 0 |
| DP [MFLOP/s] | 1059.4978 |
| AVX DP [MFLOP/s] | 1.3335 |
| Packed [MUOPS/s] | 0.1667 |
| Scalar [MUOPS/s] | 1058.1643 |
| Memory read bandwidth [MBytes/s] | 136.3006 |
| Memory read data volume [GBytes] | 0.8038 |
| Memory write bandwidth [MBytes/s] | 72.2612 |
| Memory write data volume [GBytes] | 0.4262 |
| Memory bandwidth [MBytes/s] | 208.5618 |
| Memory data volume [GBytes] | 1.2300 |
| Operational intensity | 5.0800 |
+-----------------------------------+------------+

88
likwid-outputs/csx-lammps-sp-mem_sp-stub.out Normal file
View File

@ -0,0 +1,88 @@
--------------------------------------------------------------------------------
CPU name: Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz
CPU type: Intel Cascadelake SP processor
CPU clock: 2.49 GHz
--------------------------------------------------------------------------------
Initializing parameters...
Initializing atoms...
Creating atoms...
Pattern: seq
Number of timesteps: 200
Number of atoms: 256
Number of neighbors per atom: 1024
Number of times to replicate neighbor lists: 1
Estimated total data volume (kB): 1056.7680
Estimated atom data volume (kB): 3.0720
Estimated neighborlist data volume (kB): 1050.6240
Initializing neighbor lists...
Creating neighbor lists...
Computing forces...
Total time: 0.2466, Mega atom updates/s: 0.2076
Cycles per atom: 9631.9934, Cycles per neighbor: 9.4062
Statistics:
Vector width: 16, Processor frequency: 2.0000 GHz
Average neighbors per atom: 1018.9055
Average SIMD iterations per atom: 63.6816
Total number of computed pair interactions: 52428800
Total number of SIMD iterations: 3276800
Useful read data volume for force computation: 0.84GB
Cycles/SIMD iteration: 150.4999
--------------------------------------------------------------------------------
Region force, Group 1: MEM_SP
+-------------------+------------+
| Region Info | HWThread 0 |
+-------------------+------------+
| RDTSC Runtime [s] | 0.085843 |
| call count | 200 |
+-------------------+------------+
+------------------------------------------+---------+------------+
| Event | Counter | HWThread 0 |
+------------------------------------------+---------+------------+
| INSTR_RETIRED_ANY | FIXC0 | 129769100 |
| CPU_CLK_UNHALTED_CORE | FIXC1 | 172300100 |
| CPU_CLK_UNHALTED_REF | FIXC2 | 215371300 |
| PWR_PKG_ENERGY | PWR0 | 9.2849 |
| PWR_DRAM_ENERGY | PWR3 | 0 |
| FP_ARITH_INST_RETIRED_128B_PACKED_SINGLE | PMC0 | 0 |
| FP_ARITH_INST_RETIRED_SCALAR_SINGLE | PMC1 | 154000 |
| FP_ARITH_INST_RETIRED_256B_PACKED_SINGLE | PMC2 | 0 |
| FP_ARITH_INST_RETIRED_512B_PACKED_SINGLE | PMC3 | 89088000 |
| CAS_COUNT_RD | MBOX0C0 | 8354 |
| CAS_COUNT_WR | MBOX0C1 | 1126 |
| CAS_COUNT_RD | MBOX1C0 | 7863 |
| CAS_COUNT_WR | MBOX1C1 | 1105 |
| CAS_COUNT_RD | MBOX2C0 | 7990 |
| CAS_COUNT_WR | MBOX2C1 | 1113 |
| CAS_COUNT_RD | MBOX3C0 | 4775 |
| CAS_COUNT_WR | MBOX3C1 | 1112 |
| CAS_COUNT_RD | MBOX4C0 | 4201 |
| CAS_COUNT_WR | MBOX4C1 | 1127 |
| CAS_COUNT_RD | MBOX5C0 | 4035 |
| CAS_COUNT_WR | MBOX5C1 | 1120 |
+------------------------------------------+---------+------------+
+-----------------------------------+------------+
| Metric | HWThread 0 |
+-----------------------------------+------------+
| Runtime (RDTSC) [s] | 0.0858 |
| Runtime unhalted [s] | 0.0691 |
| Clock [MHz] | 1995.2787 |
| CPI | 1.3277 |
| Energy [J] | 9.2849 |
| Power [W] | 108.1610 |
| Energy DRAM [J] | 0 |
| Power DRAM [W] | 0 |
| SP [MFLOP/s] | 16606.5397 |
| AVX SP [MFLOP/s] | 16604.7458 |
| Packed [MUOPS/s] | 1037.7966 |
| Scalar [MUOPS/s] | 1.7940 |
| Memory read bandwidth [MBytes/s] | 27.7476 |
| Memory read data volume [GBytes] | 0.0024 |
| Memory write bandwidth [MBytes/s] | 4.9974 |
| Memory write data volume [GBytes] | 0.0004 |
| Memory bandwidth [MBytes/s] | 32.7450 |
| Memory data volume [GBytes] | 0.0028 |
| Operational intensity | 507.1471 |
+-----------------------------------+------------+

168
likwid-outputs/csx-lammps-sp-mem_sp.out Normal file
View File

@ -0,0 +1,168 @@
--------------------------------------------------------------------------------
CPU name: Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz
CPU type: Intel Cascadelake SP processor
CPU clock: 2.49 GHz
--------------------------------------------------------------------------------
Parameters:
Force field: lj
Kernel: plain-C
Data layout: AoS
Floating-point precision: single
Unit cells (nx, ny, nz): 32, 32, 32
Domain box sizes (x, y, z): 5.374708e+01, 5.374708e+01, 5.374708e+01
Periodic (x, y, z): 1, 1, 1
Lattice size: 1.679596e+00
Epsilon: 1.000000e+00
Sigma: 1.000000e+00
Spring constant: 1.000000e+00
Damping constant: 1.000000e+00
Temperature: 1.440000e+00
RHO: 8.442000e-01
Mass: 1.000000e+00
Number of types: 4
Number of timesteps: 200
Report stats every (timesteps): 100
Reneighbor every (timesteps): 20
Prune every (timesteps): 1000
Output positions every (timesteps): 20
Output velocities every (timesteps): 5
Delta time (dt): 5.000000e-03
Cutoff radius: 2.500000e+00
Skin: 3.000000e-01
Half neighbor lists: 0
Processor frequency (GHz): 2.0000
----------------------------------------------------------------------------
step temp pressure
0 1.440000e+00 1.215639e+00
100 8.200897e-01 6.923144e-01
200 7.961481e-01 6.721031e-01
----------------------------------------------------------------------------
System: 131072 atoms 47265 ghost atoms, Steps: 200
TOTAL 10.83s FORCE 4.62s NEIGH 5.94s REST 0.26s
----------------------------------------------------------------------------
Performance: 2.42 million atom updates per second
Statistics:
Vector width: 16, Processor frequency: 2.0000 GHz
Average neighbors per atom: 76.0351
Average SIMD iterations per atom: 5.0875
Total number of computed pair interactions: 2003181259
Total number of SIMD iterations: 134032075
Useful read data volume for force computation: 32.79GB
Cycles/SIMD iteration: 68.9511
--------------------------------------------------------------------------------
Region force, Group 1: MEM_SP
+-------------------+------------+
| Region Info | HWThread 0 |
+-------------------+------------+
| RDTSC Runtime [s] | 4.452877 |
| call count | 201 |
+-------------------+------------+
+------------------------------------------+---------+-------------+
| Event | Counter | HWThread 0 |
+------------------------------------------+---------+-------------+
| INSTR_RETIRED_ANY | FIXC0 | 7428719000 |
| CPU_CLK_UNHALTED_CORE | FIXC1 | 8875251000 |
| CPU_CLK_UNHALTED_REF | FIXC2 | 11094050000 |
| PWR_PKG_ENERGY | PWR0 | 265.5057 |
| PWR_DRAM_ENERGY | PWR3 | 0 |
| FP_ARITH_INST_RETIRED_128B_PACKED_SINGLE | PMC0 | 0 |
| FP_ARITH_INST_RETIRED_SCALAR_SINGLE | PMC1 | 79036820 |
| FP_ARITH_INST_RETIRED_256B_PACKED_SINGLE | PMC2 | 0 |
| FP_ARITH_INST_RETIRED_512B_PACKED_SINGLE | PMC3 | 3935012000 |
| CAS_COUNT_RD | MBOX0C0 | 19716700 |
| CAS_COUNT_WR | MBOX0C1 | 595747 |
| CAS_COUNT_RD | MBOX1C0 | 19734880 |
| CAS_COUNT_WR | MBOX1C1 | 597090 |
| CAS_COUNT_RD | MBOX2C0 | 19732800 |
| CAS_COUNT_WR | MBOX2C1 | 595219 |
| CAS_COUNT_RD | MBOX3C0 | 19886430 |
| CAS_COUNT_WR | MBOX3C1 | 632443 |
| CAS_COUNT_RD | MBOX4C0 | 19887210 |
| CAS_COUNT_WR | MBOX4C1 | 633169 |
| CAS_COUNT_RD | MBOX5C0 | 19935560 |
| CAS_COUNT_WR | MBOX5C1 | 634112 |
+------------------------------------------+---------+-------------+
+-----------------------------------+------------+
| Metric | HWThread 0 |
+-----------------------------------+------------+
| Runtime (RDTSC) [s] | 4.4529 |
| Runtime unhalted [s] | 3.5585 |
| Clock [MHz] | 1995.2693 |
| CPI | 1.1947 |
| Energy [J] | 265.5057 |
| Power [W] | 59.6257 |
| Energy DRAM [J] | 0 |
| Power DRAM [W] | 0 |
| SP [MFLOP/s] | 14156.9661 |
| AVX SP [MFLOP/s] | 14139.2165 |
| Packed [MUOPS/s] | 883.7010 |
| Scalar [MUOPS/s] | 17.7496 |
| Memory read bandwidth [MBytes/s] | 1708.8254 |
| Memory read data volume [GBytes] | 7.6092 |
| Memory write bandwidth [MBytes/s] | 53.0035 |
| Memory write data volume [GBytes] | 0.2360 |
| Memory bandwidth [MBytes/s] | 1761.8288 |
| Memory data volume [GBytes] | 7.8452 |
| Operational intensity | 8.0354 |
+-----------------------------------+------------+
Region reneighbour, Group 1: MEM_SP
+-------------------+------------+
| Region Info | HWThread 0 |
+-------------------+------------+
| RDTSC Runtime [s] | 5.935627 |
| call count | 10 |
+-------------------+------------+
+------------------------------------------+---------+-------------+
| Event | Counter | HWThread 0 |
+------------------------------------------+---------+-------------+
| INSTR_RETIRED_ANY | FIXC0 | 18208530000 |
| CPU_CLK_UNHALTED_CORE | FIXC1 | 11805500000 |
| CPU_CLK_UNHALTED_REF | FIXC2 | 14756870000 |
| PWR_PKG_ENERGY | PWR0 | 340.7903 |
| PWR_DRAM_ENERGY | PWR3 | 0 |
| FP_ARITH_INST_RETIRED_128B_PACKED_SINGLE | PMC0 | 0 |
| FP_ARITH_INST_RETIRED_SCALAR_SINGLE | PMC1 | 6240406000 |
| FP_ARITH_INST_RETIRED_256B_PACKED_SINGLE | PMC2 | 0 |
| FP_ARITH_INST_RETIRED_512B_PACKED_SINGLE | PMC3 | 491520 |
| CAS_COUNT_RD | MBOX0C0 | 1772377 |
| CAS_COUNT_WR | MBOX0C1 | 975760 |
| CAS_COUNT_RD | MBOX1C0 | 1770611 |
| CAS_COUNT_WR | MBOX1C1 | 977433 |
| CAS_COUNT_RD | MBOX2C0 | 1771722 |
| CAS_COUNT_WR | MBOX2C1 | 979122 |
| CAS_COUNT_RD | MBOX3C0 | 1782901 |
| CAS_COUNT_WR | MBOX3C1 | 967621 |
| CAS_COUNT_RD | MBOX4C0 | 1780789 |
| CAS_COUNT_WR | MBOX4C1 | 967179 |
| CAS_COUNT_RD | MBOX5C0 | 1784733 |
| CAS_COUNT_WR | MBOX5C1 | 969349 |
+------------------------------------------+---------+-------------+
+-----------------------------------+------------+
| Metric | HWThread 0 |
+-----------------------------------+------------+
| Runtime (RDTSC) [s] | 5.9356 |
| Runtime unhalted [s] | 4.7334 |
| Clock [MHz] | 1995.2675 |
| CPI | 0.6483 |
| Energy [J] | 340.7903 |
| Power [W] | 57.4144 |
| Energy DRAM [J] | 0 |
| Power DRAM [W] | 0 |
| SP [MFLOP/s] | 1052.6723 |
| AVX SP [MFLOP/s] | 1.3249 |
| Packed [MUOPS/s] | 0.0828 |
| Scalar [MUOPS/s] | 1051.3474 |
| Memory read bandwidth [MBytes/s] | 114.9736 |
| Memory read data volume [GBytes] | 0.6824 |
| Memory write bandwidth [MBytes/s] | 62.9308 |
| Memory write data volume [GBytes] | 0.3735 |
| Memory bandwidth [MBytes/s] | 177.9044 |
| Memory data volume [GBytes] | 1.0560 |
| Operational intensity | 5.9171 |
+-----------------------------------+------------+

1474
static_analysis/jan/icx-icc-lammps-avx2.s Normal file
View File

File diff suppressed because it is too large Load Diff

1659
static_analysis/jan/icx-icc-lammps-avx512.s Normal file
View File

File diff suppressed because it is too large Load Diff

1310
static_analysis/jan/icx-icc-lammps-novec.s Normal file
View File

File diff suppressed because it is too large Load Diff

1522
static_analysis/jan/icx-icc-lammps-sse.s Normal file
View File

File diff suppressed because it is too large Load Diff

2103
static_analysis/jan/icx-icx-gromacs-avx512-sp.s Normal file
View File

File diff suppressed because it is too large Load Diff

2453
static_analysis/jan/icx-icx-gromacs-avx512.s Normal file
View File

File diff suppressed because it is too large Load Diff

676
static_analysis/jan/zen-icx-lammps-avx2.s Normal file
View File

@ -0,0 +1,676 @@
.text
.file "force_lj.c"
.section .rodata.cst8,"aM",@progbits,8
.p2align 3 # -- Begin function computeForceLJFullNeigh_plain_c
.LCPI0_0:
.quad 4631952216750555136 # 48
.LCPI0_3:
.quad 4607182418800017408 # 1
.LCPI0_4:
.quad -4620693217682128896 # -0.5
.section .rodata.cst4,"aM",@progbits,4
.p2align 2
.LCPI0_1:
.long 3 # 0x3
.LCPI0_2:
.long 2 # 0x2
.section .rodata.cst16,"aM",@progbits,16
.p2align 4
.LCPI0_5:
.zero 16,255
.text
.globl computeForceLJFullNeigh_plain_c
.p2align 4, 0x90
.type computeForceLJFullNeigh_plain_c,@function
computeForceLJFullNeigh_plain_c: #
.LcomputeForceLJFullNeigh_plain_c$local:
.cfi_startproc
# %bb.0: #
pushq %rbp
.cfi_def_cfa_offset 16
pushq %r15
.cfi_def_cfa_offset 24
pushq %r14
.cfi_def_cfa_offset 32
pushq %r13
.cfi_def_cfa_offset 40
pushq %r12
.cfi_def_cfa_offset 48
pushq %rbx
.cfi_def_cfa_offset 56
subq $264, %rsp # imm = 0x108
.cfi_def_cfa_offset 320
.cfi_offset %rbx, -56
.cfi_offset %r12, -48
.cfi_offset %r13, -40
.cfi_offset %r14, -32
.cfi_offset %r15, -24
.cfi_offset %rbp, -16
movq %rcx, %rbx
movq %rdx, %r15
movq %rsi, %r12
movl 4(%rsi), %r14d
vmovsd 144(%rdi), %xmm0 # xmm0 = mem[0],zero
vmovsd %xmm0, (%rsp) # 8-byte Spill
vmovsd 40(%rdi), %xmm0 # xmm0 = mem[0],zero
vmovsd %xmm0, 128(%rsp) # 8-byte Spill
vmovq 56(%rdi), %xmm0 # xmm0 = mem[0],zero
vmovdqa %xmm0, 80(%rsp) # 16-byte Spill
testl %r14d, %r14d
jle .LBB0_2
# %bb.1: #
movq 64(%r12), %rdi
leaq (,%r14,8), %rax
leaq (%rax,%rax,2), %rdx
xorl %esi, %esi
callq _intel_fast_memset
.LBB0_2: #
xorl %eax, %eax
callq getTimeStamp
vmovq %xmm0, 32(%rsp) # 8-byte Folded Spill
movl $.L.str, %edi
callq likwid_markerStartRegion
testl %r14d, %r14d
jle .LBB0_19
# %bb.3: #
vmovsd (%rsp), %xmm0 # 8-byte Reload
# xmm0 = mem[0],zero
vmulsd %xmm0, %xmm0, %xmm13
movq 16(%r15), %r11
movq 24(%r15), %rsi
movslq 8(%r15), %rdi
movq 16(%r12), %r15
movq 64(%r12), %r8
vmovsd 128(%rsp), %xmm0 # 8-byte Reload
# xmm0 = mem[0],zero
vmulsd .LCPI0_0(%rip), %xmm0, %xmm15
movq %rbx, 24(%rsp) # 8-byte Spill
vmovdqu (%rbx), %xmm14
decq %r14
vmovq %r15, %xmm0
vpbroadcastq %xmm0, %ymm3
vbroadcastsd %xmm13, %ymm2
vmovapd 80(%rsp), %xmm12 # 16-byte Reload
vbroadcastsd %xmm12, %ymm8
vbroadcastsd %xmm15, %ymm9
shlq $2, %rdi
xorl %r10d, %r10d
movq %r14, 56(%rsp) # 8-byte Spill
vmovapd %xmm13, 192(%rsp) # 16-byte Spill
movq %rsi, 48(%rsp) # 8-byte Spill
movq %rdi, 40(%rsp) # 8-byte Spill
vmovapd %xmm15, 176(%rsp) # 16-byte Spill
vmovupd %ymm2, 224(%rsp) # 32-byte Spill
vmovupd %ymm9, 128(%rsp) # 32-byte Spill
jmp .LBB0_6
.p2align 4, 0x90
.LBB0_17: #
# in Loop: Header=BB0_6 Depth=1
movq %r13, %rdx
.LBB0_5: #
# in Loop: Header=BB0_6 Depth=1
vaddsd (%r8,%r12,8), %xmm10, %xmm0
vmovsd %xmm0, (%r8,%r12,8)
vaddsd (%r8,%rbx,8), %xmm11, %xmm0
vmovsd %xmm0, (%r8,%rbx,8)
vaddsd (%r8,%rbp,8), %xmm5, %xmm0
vmovsd %xmm0, (%r8,%rbp,8)
leal 3(%r13), %eax
addl $6, %r13d
testl %eax, %eax
cmovnsl %eax, %r13d
sarl $2, %r13d
movslq %r13d, %rax
vmovq %rax, %xmm0
vmovq %rdx, %xmm1
vpunpcklqdq %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0]
vpaddq %xmm0, %xmm14, %xmm14
addq %rdi, %r11
cmpq %r14, %r10
leaq 1(%r10), %r10
je .LBB0_18
.LBB0_6: #
# =>This Loop Header: Depth=1
# Child Loop BB0_9 Depth 2
# Child Loop BB0_13 Depth 2
movl (%rsi,%r10,4), %r13d
leal (%r10,%r10,2), %r12d
leal (%r10,%r10,2), %ebx
incl %ebx
leal (%r10,%r10,2), %ebp
addl $2, %ebp
testl %r13d, %r13d
jle .LBB0_4
# %bb.7: #
# in Loop: Header=BB0_6 Depth=1
vmovsd (%r15,%r12,8), %xmm0 # xmm0 = mem[0],zero
vmovsd (%r15,%rbx,8), %xmm1 # xmm1 = mem[0],zero
vmovsd (%r15,%rbp,8), %xmm2 # xmm2 = mem[0],zero
movq %r13, %rdx
movl $4294967292, %eax # imm = 0xFFFFFFFC
andq %rax, %rdx
vmovapd %xmm0, 112(%rsp) # 16-byte Spill
vmovapd %xmm1, 96(%rsp) # 16-byte Spill
vmovapd %xmm2, (%rsp) # 16-byte Spill
je .LBB0_16
# %bb.8: #
# in Loop: Header=BB0_6 Depth=1
movq %rbp, 64(%rsp) # 8-byte Spill
movq %rbx, 72(%rsp) # 8-byte Spill
vmovdqa %xmm14, 208(%rsp) # 16-byte Spill
vbroadcastsd %xmm0, %ymm14
vbroadcastsd %xmm1, %ymm5
vbroadcastsd %xmm2, %ymm10
vxorpd %xmm0, %xmm0, %xmm0
vxorpd %xmm15, %xmm15, %xmm15
vxorpd %xmm13, %xmm13, %xmm13
xorl %ebp, %ebp
vmovapd %ymm8, %ymm9
vmovupd 224(%rsp), %ymm8 # 32-byte Reload
.p2align 4, 0x90
movl $111, %ebx # OSACA START MARKER
.byte 100 # OSACA START MARKER
.byte 103 # OSACA START MARKER
.byte 144 # OSACA START MARKER
# LLVM-MCA-BEGIN
# pointer_increment=16 e95035fc9e97f63299dd5188a0872bfc
.LBB0_9: #
# Parent Loop BB0_6 Depth=1
# => This Inner Loop Header: Depth=2
vpbroadcastd .LCPI0_1(%rip), %xmm1 # xmm1 = [3,3,3,3]
vpmulld (%r11,%rbp,4), %xmm1, %xmm11
vpmovsxdq %xmm11, %ymm1
vpsllq $3, %ymm1, %ymm1
vpaddq %ymm1, %ymm3, %ymm1
vmovq %xmm1, %r14
vpextrq $1, %xmm1, %r9
vextracti128 $1, %ymm1, %xmm1
vmovsd (%r14), %xmm2 # xmm2 = mem[0],zero
vpsubd .LCPI0_5, %xmm11, %xmm6
vpmovsxdq %xmm6, %ymm6
vpsllq $3, %ymm6, %ymm6
vmovq %xmm1, %rdi
vpaddq %ymm6, %ymm3, %ymm6
vmovq %xmm6, %rcx
vpextrq $1, %xmm1, %rbx
vpextrq $1, %xmm6, %rax
vextracti128 $1, %ymm6, %xmm1
vmovsd (%rdi), %xmm6 # xmm6 = mem[0],zero
vmovq %xmm1, %rdi
vpextrq $1, %xmm1, %rsi
vmovsd (%rdi), %xmm1 # xmm1 = mem[0],zero
vmovsd (%rcx), %xmm7 # xmm7 = mem[0],zero
vpbroadcastd .LCPI0_2(%rip), %xmm12 # xmm12 = [2,2,2,2]
vmovhpd (%r9), %xmm2, %xmm2 # xmm2 = xmm2[0],mem[0]
vpaddd %xmm12, %xmm11, %xmm4
vpmovsxdq %xmm4, %ymm4
vmovhpd (%rax), %xmm7, %xmm7 # xmm7 = xmm7[0],mem[0]
vpsllq $3, %ymm4, %ymm4
vpaddq %ymm4, %ymm3, %ymm4
vmovhpd (%rbx), %xmm6, %xmm6 # xmm6 = xmm6[0],mem[0]
vpextrq $1, %xmm4, %rax
vmovhpd (%rsi), %xmm1, %xmm1 # xmm1 = xmm1[0],mem[0]
vmovq %xmm4, %rcx
vextracti128 $1, %ymm4, %xmm4
vmovq %xmm4, %rsi
vinsertf128 $1, %xmm6, %ymm2, %ymm2
vpextrq $1, %xmm4, %rdi
vmovsd (%rsi), %xmm4 # xmm4 = mem[0],zero
vsubpd %ymm2, %ymm14, %ymm2
vmovhpd (%rdi), %xmm4, %xmm4 # xmm4 = xmm4[0],mem[0]
vmovsd (%rcx), %xmm6 # xmm6 = mem[0],zero
vinsertf128 $1, %xmm1, %ymm7, %ymm1
vmovhpd (%rax), %xmm6, %xmm6 # xmm6 = xmm6[0],mem[0]
vinsertf128 $1, %xmm4, %ymm6, %ymm4
vsubpd %ymm1, %ymm5, %ymm1
vsubpd %ymm4, %ymm10, %ymm4
vmulpd %ymm2, %ymm2, %ymm6
vfmadd231pd %ymm1, %ymm1, %ymm6 # ymm6 = (ymm1 * ymm1) + ymm6
vfmadd231pd %ymm4, %ymm4, %ymm6 # ymm6 = (ymm4 * ymm4) + ymm6
vbroadcastsd .LCPI0_3(%rip), %ymm7 # ymm7 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0]
vdivpd %ymm6, %ymm7, %ymm7
vmulpd %ymm7, %ymm7, %ymm11
vmulpd %ymm9, %ymm11, %ymm11
vbroadcastsd .LCPI0_4(%rip), %ymm12 # ymm12 = [-5.0E-1,-5.0E-1,-5.0E-1,-5.0E-1]
vmulpd %ymm7, %ymm11, %ymm11
vaddpd %ymm12, %ymm11, %ymm12
vmulpd 128(%rsp), %ymm7, %ymm7 # 32-byte Folded Reload
vmulpd %ymm7, %ymm11, %ymm7
vmulpd %ymm7, %ymm12, %ymm7
vcmpltpd %ymm8, %ymm6, %ymm6
vfmadd213pd %ymm0, %ymm7, %ymm2 # ymm2 = (ymm7 * ymm2) + ymm0
vblendvpd %ymm6, %ymm2, %ymm0, %ymm0
vfmadd213pd %ymm15, %ymm7, %ymm1 # ymm1 = (ymm7 * ymm1) + ymm15
vfmadd213pd %ymm13, %ymm7, %ymm4 # ymm4 = (ymm7 * ymm4) + ymm13
vblendvpd %ymm6, %ymm1, %ymm15, %ymm15
vblendvpd %ymm6, %ymm4, %ymm13, %ymm13
addq $4, %rbp
cmpq %rdx, %rbp
jb .LBB0_9
# LLVM-MCA-END
movl $222, %ebx # OSACA END MARKER
.byte 100 # OSACA END MARKER
.byte 103 # OSACA END MARKER
.byte 144 # OSACA END MARKER
# %bb.10: #
# in Loop: Header=BB0_6 Depth=1
vpermilpd $1, %xmm0, %xmm1 # xmm1 = xmm0[1,0]
vaddsd %xmm1, %xmm0, %xmm1
vextractf128 $1, %ymm0, %xmm0
vaddsd %xmm0, %xmm1, %xmm1
vpermilpd $1, %xmm0, %xmm0 # xmm0 = xmm0[1,0]
vaddsd %xmm0, %xmm1, %xmm10
vpermilpd $1, %xmm15, %xmm1 # xmm1 = xmm15[1,0]
vaddsd %xmm1, %xmm15, %xmm1
vextractf128 $1, %ymm15, %xmm2
vaddsd %xmm2, %xmm1, %xmm1
vpermilpd $1, %xmm2, %xmm2 # xmm2 = xmm2[1,0]
vaddsd %xmm2, %xmm1, %xmm11
vpermilpd $1, %xmm13, %xmm1 # xmm1 = xmm13[1,0]
vaddsd %xmm1, %xmm13, %xmm1
vextractf128 $1, %ymm13, %xmm2
vaddsd %xmm2, %xmm1, %xmm1
vpermilpd $1, %xmm2, %xmm2 # xmm2 = xmm2[1,0]
vaddsd %xmm2, %xmm1, %xmm5
movq 56(%rsp), %r14 # 8-byte Reload
vmovapd 80(%rsp), %xmm12 # 16-byte Reload
vmovapd 192(%rsp), %xmm13 # 16-byte Reload
movq 48(%rsp), %rsi # 8-byte Reload
movq 40(%rsp), %rdi # 8-byte Reload
vmovdqa 208(%rsp), %xmm14 # 16-byte Reload
vmovapd 176(%rsp), %xmm15 # 16-byte Reload
vmovapd %ymm9, %ymm8
movq 72(%rsp), %rbx # 8-byte Reload
movq 64(%rsp), %rbp # 8-byte Reload
vmovapd 112(%rsp), %xmm0 # 16-byte Reload
cmpq %r13, %rdx
jae .LBB0_17
jmp .LBB0_11
.p2align 4, 0x90
.LBB0_4: #
# in Loop: Header=BB0_6 Depth=1
movslq %r13d, %rdx
vxorpd %xmm5, %xmm5, %xmm5
vxorpd %xmm11, %xmm11, %xmm11
vxorpd %xmm10, %xmm10, %xmm10
jmp .LBB0_5
.p2align 4, 0x90
.LBB0_16: #
# in Loop: Header=BB0_6 Depth=1
vxorpd %xmm10, %xmm10, %xmm10
vxorpd %xmm11, %xmm11, %xmm11
vxorpd %xmm5, %xmm5, %xmm5
cmpq %r13, %rdx
jae .LBB0_17
.LBB0_11: #
# in Loop: Header=BB0_6 Depth=1
vmovapd 96(%rsp), %xmm4 # 16-byte Reload
jmp .LBB0_13
.p2align 4, 0x90
.LBB0_12: #
# in Loop: Header=BB0_13 Depth=2
incq %rdx
cmpq %rdx, %r13
je .LBB0_17
.LBB0_13: #
# Parent Loop BB0_6 Depth=1
# => This Inner Loop Header: Depth=2
movl (%r11,%rdx,4), %eax
leal (%rax,%rax,2), %ecx
movslq %ecx, %rcx
vsubsd (%r15,%rcx,8), %xmm0, %xmm6
leal (%rax,%rax,2), %ecx
incl %ecx
movslq %ecx, %rcx
vsubsd (%r15,%rcx,8), %xmm4, %xmm2
leal 2(%rax,%rax,2), %eax
cltq
vmovapd (%rsp), %xmm1 # 16-byte Reload
vsubsd (%r15,%rax,8), %xmm1, %xmm1
vmulsd %xmm6, %xmm6, %xmm7
vfmadd231sd %xmm2, %xmm2, %xmm7 # xmm7 = (xmm2 * xmm2) + xmm7
vfmadd231sd %xmm1, %xmm1, %xmm7 # xmm7 = (xmm1 * xmm1) + xmm7
vucomisd %xmm13, %xmm7
jae .LBB0_12
# %bb.14: #
# in Loop: Header=BB0_13 Depth=2
vmovsd .LCPI0_3(%rip), %xmm0 # xmm0 = mem[0],zero
vdivsd %xmm7, %xmm0, %xmm7
vmulsd %xmm7, %xmm7, %xmm0
vmulsd %xmm0, %xmm12, %xmm0
vmulsd %xmm7, %xmm0, %xmm0
vaddsd .LCPI0_4(%rip), %xmm0, %xmm4
vmulsd %xmm7, %xmm15, %xmm7
vmulsd %xmm0, %xmm7, %xmm0
vmulsd %xmm4, %xmm0, %xmm0
vmovapd 96(%rsp), %xmm4 # 16-byte Reload
vfmadd231sd %xmm6, %xmm0, %xmm10 # xmm10 = (xmm0 * xmm6) + xmm10
vfmadd231sd %xmm2, %xmm0, %xmm11 # xmm11 = (xmm0 * xmm2) + xmm11
vfmadd231sd %xmm1, %xmm0, %xmm5 # xmm5 = (xmm0 * xmm1) + xmm5
vmovapd 112(%rsp), %xmm0 # 16-byte Reload
jmp .LBB0_12
.LBB0_18: #
movq 24(%rsp), %rax # 8-byte Reload
vmovdqu %xmm14, (%rax)
.LBB0_19: #
movl $.L.str, %edi
vzeroupper
callq likwid_markerStopRegion
xorl %eax, %eax
callq getTimeStamp
vsubsd 32(%rsp), %xmm0, %xmm0 # 8-byte Folded Reload
addq $264, %rsp # imm = 0x108
.cfi_def_cfa_offset 56
popq %rbx
.cfi_def_cfa_offset 48
popq %r12
.cfi_def_cfa_offset 40
popq %r13
.cfi_def_cfa_offset 32
popq %r14
.cfi_def_cfa_offset 24
popq %r15
.cfi_def_cfa_offset 16
popq %rbp
.cfi_def_cfa_offset 8
retq
.Lfunc_end0:
.size computeForceLJFullNeigh_plain_c, .Lfunc_end0-computeForceLJFullNeigh_plain_c
.cfi_endproc
# -- End function
.section .rodata.cst8,"aM",@progbits,8
.p2align 3 # -- Begin function computeForceLJHalfNeigh
.LCPI1_0:
.quad 4631952216750555136 # 48
.LCPI1_1:
.quad 4607182418800017408 # 1
.LCPI1_2:
.quad -4620693217682128896 # -0.5
.LCPI1_3:
.quad 4741671816366391296 # 1.0E+9
.text
.globl computeForceLJHalfNeigh
.p2align 4, 0x90
.type computeForceLJHalfNeigh,@function
computeForceLJHalfNeigh: #
.LcomputeForceLJHalfNeigh$local:
.cfi_startproc
# %bb.0: #
pushq %rbp
.cfi_def_cfa_offset 16
pushq %r15
.cfi_def_cfa_offset 24
pushq %r14
.cfi_def_cfa_offset 32
pushq %r13
.cfi_def_cfa_offset 40
pushq %r12
.cfi_def_cfa_offset 48
pushq %rbx
.cfi_def_cfa_offset 56
subq $56, %rsp
.cfi_def_cfa_offset 112
.cfi_offset %rbx, -56
.cfi_offset %r12, -48
.cfi_offset %r13, -40
.cfi_offset %r14, -32
.cfi_offset %r15, -24
.cfi_offset %rbp, -16
movq %rcx, 24(%rsp) # 8-byte Spill
movq %rdx, %r12
movq %rsi, %r13
movl 4(%rsi), %r15d
vmovsd 144(%rdi), %xmm0 # xmm0 = mem[0],zero
vmovsd %xmm0, (%rsp) # 8-byte Spill
vmovsd 40(%rdi), %xmm0 # xmm0 = mem[0],zero
vmovsd %xmm0, 8(%rsp) # 8-byte Spill
movq %rdi, 40(%rsp) # 8-byte Spill
vmovsd 56(%rdi), %xmm0 # xmm0 = mem[0],zero
vmovsd %xmm0, 16(%rsp) # 8-byte Spill
testl %r15d, %r15d
jle .LBB1_2
# %bb.1: #
movq 64(%r13), %rdi
leaq (,%r15,8), %rax
leaq (%rax,%rax,2), %rdx
xorl %esi, %esi
callq _intel_fast_memset
.LBB1_2: #
xorl %r14d, %r14d
xorl %eax, %eax
callq getTimeStamp
vmovsd %xmm0, 32(%rsp) # 8-byte Spill
movl $.L.str.1, %edi
callq likwid_markerStartRegion
testl %r15d, %r15d
jle .LBB1_8
# %bb.3: #
vmovsd (%rsp), %xmm0 # 8-byte Reload
# xmm0 = mem[0],zero
vmulsd %xmm0, %xmm0, %xmm12
movq 16(%r12), %rax
movq 24(%r12), %rcx
movq %rcx, (%rsp) # 8-byte Spill
movslq 8(%r12), %rdx
movq 16(%r13), %rsi
movq 64(%r13), %rdi
vmovsd 8(%rsp), %xmm0 # 8-byte Reload
# xmm0 = mem[0],zero
vmulsd .LCPI1_0(%rip), %xmm0, %xmm11
movq 24(%rsp), %rcx # 8-byte Reload
vmovdqu (%rcx), %xmm10
shlq $2, %rdx
movq %rdx, 48(%rsp) # 8-byte Spill
xorl %r13d, %r13d
xorl %r14d, %r14d
jmp .LBB1_4
.p2align 4, 0x90
.LBB1_14: #
# in Loop: Header=BB1_4 Depth=1
movq 8(%rsp), %rbp # 8-byte Reload
.LBB1_6: #
# in Loop: Header=BB1_4 Depth=1
addl %r10d, %r14d
vaddsd (%rdi,%r12,8), %xmm14, %xmm0
vmovsd %xmm0, (%rdi,%r12,8)
vaddsd (%rdi,%rbp,8), %xmm15, %xmm0
vmovsd %xmm0, (%rdi,%rbp,8)
vaddsd (%rdi,%r11,8), %xmm13, %xmm0
vmovsd %xmm0, (%rdi,%r11,8)
leal 3(%r10), %ecx
addl $6, %r10d
testl %ecx, %ecx
cmovnsl %ecx, %r10d
sarl $2, %r10d
movslq %r10d, %rcx
vmovq %rcx, %xmm0
vmovq %rdx, %xmm1
vpunpcklqdq %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0]
vpaddq %xmm0, %xmm10, %xmm10
incq %r13
addq 48(%rsp), %rax # 8-byte Folded Reload
cmpq %r15, %r13
je .LBB1_7
.LBB1_4: #
# =>This Loop Header: Depth=1
# Child Loop BB1_10 Depth 2
movq (%rsp), %rcx # 8-byte Reload
movslq (%rcx,%r13,4), %r10
leaq (,%r13,2), %rcx
addq %r13, %rcx
leal 1(%rcx), %ebp
leal 2(%rcx), %r11d
movl %ecx, %r12d
testq %r10, %r10
jle .LBB1_5
# %bb.9: #
# in Loop: Header=BB1_4 Depth=1
vmovsd (%rsi,%r12,8), %xmm9 # xmm9 = mem[0],zero
movq %rbp, 8(%rsp) # 8-byte Spill
vmovsd (%rsi,%rbp,8), %xmm4 # xmm4 = mem[0],zero
vmovsd (%rsi,%r11,8), %xmm1 # xmm1 = mem[0],zero
movl %r10d, %edx
vxorpd %xmm14, %xmm14, %xmm14
xorl %ebx, %ebx
vxorpd %xmm15, %xmm15, %xmm15
vxorpd %xmm13, %xmm13, %xmm13
jmp .LBB1_10
.p2align 4, 0x90
.LBB1_13: #
# in Loop: Header=BB1_10 Depth=2
incq %rbx
cmpq %rbx, %rdx
je .LBB1_14
.LBB1_10: #
# Parent Loop BB1_4 Depth=1
# => This Inner Loop Header: Depth=2
movslq (%rax,%rbx,4), %r9
leaq (%r9,%r9,2), %r8
vsubsd (%rsi,%r8,8), %xmm9, %xmm2
movslq %r8d, %rcx
vsubsd 8(%rsi,%rcx,8), %xmm4, %xmm5
vsubsd 16(%rsi,%rcx,8), %xmm1, %xmm0
vmulsd %xmm2, %xmm2, %xmm6
vfmadd231sd %xmm5, %xmm5, %xmm6 # xmm6 = (xmm5 * xmm5) + xmm6
vfmadd231sd %xmm0, %xmm0, %xmm6 # xmm6 = (xmm0 * xmm0) + xmm6
vucomisd %xmm12, %xmm6
jae .LBB1_13
# %bb.11: #
# in Loop: Header=BB1_10 Depth=2
vmovsd .LCPI1_1(%rip), %xmm3 # xmm3 = mem[0],zero
vdivsd %xmm6, %xmm3, %xmm6
vmulsd 16(%rsp), %xmm6, %xmm7 # 8-byte Folded Reload
vmulsd %xmm6, %xmm6, %xmm8
vmulsd %xmm7, %xmm8, %xmm7
vaddsd .LCPI1_2(%rip), %xmm7, %xmm3
vmulsd %xmm6, %xmm11, %xmm6
vmulsd %xmm7, %xmm6, %xmm6
vmulsd %xmm3, %xmm6, %xmm3
vmulsd %xmm2, %xmm3, %xmm6
vaddsd %xmm6, %xmm14, %xmm14
vmulsd %xmm5, %xmm3, %xmm2
vaddsd %xmm2, %xmm15, %xmm15
vmulsd %xmm0, %xmm3, %xmm0
vaddsd %xmm0, %xmm13, %xmm13
cmpl %r15d, %r9d
jge .LBB1_13
# %bb.12: #
# in Loop: Header=BB1_10 Depth=2
leaq 1(%rcx), %rbp
addq $2, %rcx
vmovsd (%rdi,%r8,8), %xmm3 # xmm3 = mem[0],zero
vsubsd %xmm6, %xmm3, %xmm3
vmovsd %xmm3, (%rdi,%r8,8)
vmovsd (%rdi,%rbp,8), %xmm3 # xmm3 = mem[0],zero
vsubsd %xmm2, %xmm3, %xmm2
vmovsd %xmm2, (%rdi,%rbp,8)
vmovsd (%rdi,%rcx,8), %xmm2 # xmm2 = mem[0],zero
vsubsd %xmm0, %xmm2, %xmm0
vmovsd %xmm0, (%rdi,%rcx,8)
jmp .LBB1_13
.p2align 4, 0x90
.LBB1_5: #
# in Loop: Header=BB1_4 Depth=1
vxorpd %xmm13, %xmm13, %xmm13
movq %r10, %rdx
vxorpd %xmm15, %xmm15, %xmm15
vxorpd %xmm14, %xmm14, %xmm14
jmp .LBB1_6
.LBB1_7: #
movq 24(%rsp), %rax # 8-byte Reload
vmovdqu %xmm10, (%rax)
.LBB1_8: #
movl $.L.str.1, %edi
callq likwid_markerStopRegion
xorl %eax, %eax
callq getTimeStamp
movq 40(%rsp), %rax # 8-byte Reload
vmovsd 264(%rax), %xmm3 # xmm3 = mem[0],zero
vsubsd 32(%rsp), %xmm0, %xmm2 # 8-byte Folded Reload
vmulsd .LCPI1_3(%rip), %xmm3, %xmm0
vmulsd %xmm2, %xmm0, %xmm0
vmovapd %xmm2, %xmm1
vmovsd %xmm2, 16(%rsp) # 8-byte Spill
movl %r14d, %eax
vxorps %xmm12, %xmm12, %xmm12
vcvtsi2sd %rax, %xmm12, %xmm2
vdivsd %xmm2, %xmm0, %xmm2
movl $.L.str.2, %edi
movl %r14d, %esi
vmovapd %xmm3, %xmm0
movb $3, %al
callq printf
vmovsd 16(%rsp), %xmm0 # 8-byte Reload
# xmm0 = mem[0],zero
addq $56, %rsp
.cfi_def_cfa_offset 56
popq %rbx
.cfi_def_cfa_offset 48
popq %r12
.cfi_def_cfa_offset 40
popq %r13
.cfi_def_cfa_offset 32
popq %r14
.cfi_def_cfa_offset 24
popq %r15
.cfi_def_cfa_offset 16
popq %rbp
.cfi_def_cfa_offset 8
retq
.Lfunc_end1:
.size computeForceLJHalfNeigh, .Lfunc_end1-computeForceLJHalfNeigh
.cfi_endproc
# -- End function
.globl computeForceLJFullNeigh_simd # -- Begin function computeForceLJFullNeigh_simd
.p2align 4, 0x90
.type computeForceLJFullNeigh_simd,@function
computeForceLJFullNeigh_simd: #
.LcomputeForceLJFullNeigh_simd$local:
.cfi_startproc
# %bb.0: #
pushq %rax
.cfi_def_cfa_offset 16
movl 4(%rsi), %eax
testl %eax, %eax
jle .LBB2_2
# %bb.1: #
movq 64(%rsi), %rdi
shlq $3, %rax
leaq (%rax,%rax,2), %rdx
xorl %esi, %esi
callq _intel_fast_memset
.LBB2_2: #
xorl %eax, %eax
callq getTimeStamp
movl $.L.str, %edi
callq likwid_markerStartRegion
movq stderr(%rip), %rcx
movl $.L.str.3, %edi
movl $65, %esi
movl $1, %edx
callq fwrite
movl $-1, %edi
callq exit
.Lfunc_end2:
.size computeForceLJFullNeigh_simd, .Lfunc_end2-computeForceLJFullNeigh_simd
.cfi_endproc
# -- End function
.type .L.str,@object #
.section .rodata.str1.1,"aMS",@progbits,1
.L.str:
.asciz "force"
.size .L.str, 6
.type .L.str.1,@object #
.L.str.1:
.asciz "forceLJ-halfneigh"
.size .L.str.1, 18
.type .L.str.2,@object #
.L.str.2:
.asciz "Its: %u Freq: %f Time: %f\nCy/it: %f (half-neigh)\n"
.size .L.str.2, 52
.type .L.str.3,@object #
.L.str.3:
.asciz "Error: SIMD kernel not implemented for specified instruction set!"
.size .L.str.3, 66
.ident "Intel(R) oneAPI DPC++ Compiler 2021.1-beta05 (2020.2.0.0304)"
.section ".note.GNU-stack","",@progbits