options

Functions and Loops

35 loops and 103 functions have been discarded from the report because their ratio ((Max Inclusive Time Over Threads * 100) / Max Thread Active Time) is lower than the threshold set by object_coverage_threshold (0.1%). It represents about 0.22% of the application. To include them, change the value of object_coverage_threshold in the experiment directory configuration file, then rerun the command with the additionnal parameter --force-static-analysis.
Inclusive metrics are only related to the given object code and do not include other external objects / libraries.

Columns Filter

Max Thread Time / Walltime icx_3 (%) Coverage icx_3 (%) Coverage Excluding Loops icx_3 (%) Max Inclusive Time Over Threads icx_3 (s) Max Exclusive Time Over Threads icx_3 (s) Inclusive Time w.r.t. Wall Time icx_3 (s) Exclusive Time w.r.t. Wall Time icx_3 (s) Nb Threads icx_3 Deviation (coverage) icx_3 Deviation (walltime) icx_3 Categories icx_3 GFLOPS icx_3 Compilation Options Max Thread Time / Walltime Coverage Coverage Excluding Loops Max Inclusive Time Over Threads Max Exclusive Time Over Threads Inclusive Time w.r.t. Wall Time Exclusive Time w.r.t. Wall Time Nb Threads Deviation (coverage) Deviation (walltime) Categories GFLOPS Compilation Options
NameModuleMax Thread Time / Walltime icx_3 (%)Coverage icx_3 (%)Coverage Excluding Loops icx_3 (%)Max Inclusive Time Over Threads icx_3 (s)Max Exclusive Time Over Threads icx_3 (s)Inclusive Time w.r.t. Wall Time icx_3 (s)Exclusive Time w.r.t. Wall Time icx_3 (s)Nb Threads icx_3Deviation (coverage) icx_3Deviation (walltime) icx_3Categories icx_3GFLOPS icx_3Compilation Options
ggml_vec_dot_q8_0_q8_0+libggml-cpu.so82.6761.750.0456.930.0837.570.0219230.8418.80/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.0054.00clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_REP...
Loop 5765 - quants.c:1066-1073 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 5766 - quants.c:108-1042 - libggml-cpu.so [...]82.5861.7161.7156.8756.8737.5437.5419230.8418.8053.96
kmp_flag_64<false, true>::wait(kmp_info*, int, void*)libiomp5.so80.1835.9735.9755.2155.2121.8821.8819230.2318.02OMP (%): 100.000.11
kmp_flag_native<unsigned long long, (flag_type)1, true>::notdone_check()libiomp5.so2.411.001.001.661.660.610.611920.800.47OMP (%): 100.000.09
ggml_compute_forward_mul_mat+libggml-cpu.so0.450.220.040.310.100.130.031920.100.06/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.0042.44clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_REP...
Loop 240 - ggml-cpu.c:1316-1328 - libggml-cpu.so+0.010.000.000.030.010.000.0020.000.000.00
Loop 239 - ggml-cpu.c:1317-1328 - libggml-cpu.so0.030.000.000.020.020.000.00460.010.004.13
Loop 232 - ggml-cpu.c:1125-1395 - libggml-cpu.so [...]+0.060.170.010.380.040.100.011340.020.0145.90
Loop 234 - ggml-cpu.c:1162-1198 - libggml-cpu.so [...]+0.010.150.000.340.010.090.00140.010.00332.81
Loop 235 - ggml-cpu.c:1163-1198 - libggml-cpu.so [...]+0.120.150.020.330.080.090.011860.020.0160.92
Loop 236 - ggml-cpu.c:1164-1198 - libggml-cpu.so [...]+0.250.130.120.250.170.080.071920.070.0445.49
Loop 233 - ggml-cpu.c:1193-1194 - libggml-cpu.so0.100.010.010.070.070.010.011640.020.01123.73
Loop 237 - ggml-cpu.c:1197-1198 - libggml-cpu.so0.010.000.000.010.010.000.00600.010.00233.77
Loop 238 - ggml-cpu.c:1197-1198 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 246 - ggml-cpu.c:1248-1260 - libggml-cpu.so+0.010.010.000.030.010.000.0020.000.000.00
Loop 245 - ggml-cpu.c:1249-1260 - libggml-cpu.so0.030.010.010.020.020.000.00500.010.000.00
Loop 241 - ggml-cpu.c:1289-1295 - libggml-cpu.so+0.010.000.000.030.010.000.0010.000.000.00
Loop 242 - ggml-cpu.c:1290-1295 - libggml-cpu.so+0.010.000.000.020.010.000.0030.000.000.00
Loop 244 - ggml-cpu.c:1291-1295 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 243 - ggml-cpu.c:1291-1295 - libggml-cpu.so0.010.000.000.010.010.000.00260.000.000.00
__GI___sched_yieldlibc.so.60.480.150.150.330.330.090.091500.150.09System (%): 100.000.00
__kmp_hyper_barrier_gather(barrier_type, kmp_info*, int, int, void (*)(void*, void*), void*)libiomp5.so0.600.140.140.410.410.080.081890.200.12OMP (%): 100.0021.20
_ZN12_GLOBAL__N_115tinyBLAS_Q0_AVXI10block_q8_0S1_fE7gemm4xNILi2EEEvllll.A+libggml-cpu.so0.190.110.000.130.000.070.001890.070.04/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.00104.10clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 5325 - sgemm.cpp:814-853 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 5327 - sgemm.cpp:138-1044 - libggml-cpu.so [...]+0.000.110.000.130.000.070.0040.000.000.00
Loop 5326 - sgemm.cpp:138-1044 - libggml-cpu.so [...]0.190.110.110.130.130.070.071890.070.04103.91
_ZN12_GLOBAL__N_115tinyBLAS_Q0_AVXI10block_q8_0S1_fE7gemm4xNILi4EEEvllll.A+libggml-cpu.so0.190.110.000.130.000.070.001890.060.04/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.00224.42clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 5301 - sgemm.cpp:138-1044 - libggml-cpu.so [...]+0.000.110.000.130.000.070.0080.000.000.00
Loop 5300 - sgemm.cpp:138-1044 - libggml-cpu.so [...]0.190.110.110.130.130.070.071890.060.04224.01
Loop 5299 - sgemm.cpp:814-853 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
ggml_compute_forward_flash_attn_ext+libggml-cpu.so0.360.050.010.250.030.030.00820.140.09/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.001681.03clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 4171 - ops.cpp:8759-8927 - libggml-cpu.so [...]+0.030.050.000.360.020.030.00240.010.01116.61
Loop 4172 - vec.h:687-688 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4173 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4179 - ops.cpp:8885-8886 - libggml-cpu.so [...]0.030.000.000.020.020.000.00310.010.01358.37
Loop 4178 - ops.cpp:8885-8886 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 4176 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4174 - vec.h:677-682 - libggml-cpu.so0.010.000.000.010.010.000.00140.010.00166.38
Loop 4180 - ops.cpp:8759-8927 - libggml-cpu.so [...]+0.120.040.010.310.080.030.01330.030.024020.85
Loop 4184 - vec.h:375-381 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4188 - vec.h:502-503 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 4182 - vec.h:375-381 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4187 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4191 - vec.h:740-745 - libggml-cpu.so0.010.000.000.010.010.000.0080.010.01106.79
Loop 4185 - vec.h:687-688 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4189 - vec.h:491-497 - libggml-cpu.so0.320.030.030.220.220.020.02330.080.051486.96
Loop 4183 - vec.h:386-387 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4181 - vec.h:386-387 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4186 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4190 - vec.h:750-751 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 4177 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 4175 - vec.h:687-688 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
ggml_vec_dot_f16+libggml-cpu.so0.380.040.000.260.050.020.00330.090.05/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.00848.77clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 1776 - vec.cpp:324-325 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1777 - vec.cpp:324-325 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1778 - vec.cpp:311-316 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 1779 - vec.cpp:311-316 - libggml-cpu.so0.350.040.040.240.240.020.02330.080.05700.95
quantize_row_q8_0+libggml-cpu.so0.130.040.010.090.030.020.001640.030.02/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.000.41clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_REP...
Loop 5745 - quants.c:298-355 - libggml-cpu.so [...]0.130.030.030.090.090.020.021510.030.020.50
ggml_graph_compute_thread+libggml-cpu.so0.150.040.010.100.050.020.011790.030.02/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.0032.08clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_REP...
Loop 264 - ggml-cpu.c:2087-2088 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 263 - ggml-cpu.c:2087-2088 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 262 - ggml-cpu.c:533-2891 - libggml-cpu.so [...]0.120.020.020.080.080.010.011730.020.0150.77
__kmp_hyper_barrier_release(barrier_type, kmp_info*, int, int, int, void*)libiomp5.so0.920.040.040.630.630.020.021350.130.08OMP (%): 100.007.43
__intel_avx_rep_memcpy+exec0.130.040.040.090.090.020.021900.030.02Memory (%): 100.0067.97
Loop 4099 - - exec0.000.000.000.000.000.000.0000.000.000.00
Loop 4100 - - exec0.010.000.000.010.010.000.0060.010.009.51
__kmp_yieldlibiomp5.so0.150.040.040.100.100.020.021220.040.02OMP (%): 100.000.00
ggml_compute_forward_rope_f32+libggml-cpu.so0.170.030.000.120.020.020.001920.030.02/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.00287.54clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 3400 - ops.cpp:6210-6484 - libggml-cpu.so [...]+0.010.030.000.190.010.020.00180.010.00165.79
Loop 3401 - ops.cpp:6210-6484 - libggml-cpu.so [...]+0.060.030.000.180.040.020.001300.010.01230.83
Loop 3415 - ops.cpp:6220-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3425 - ops.cpp:6210-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3428 - ops.cpp:6220-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3420 - ops.cpp:6210-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3429 - ops.cpp:6220-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3433 - ops.cpp:6220-6245 - libggml-cpu.so [...]0.100.020.020.070.070.010.011920.020.01288.08
Loop 3412 - ops.cpp:6210-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3419 - ops.cpp:6210-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3430 - ops.cpp:6210-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3427 - ops.cpp:6210-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3414 - ops.cpp:6220-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3432 - ops.cpp:6220-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3417 - ops.cpp:6210-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3424 - ops.cpp:6220-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3423 - ops.cpp:6210-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3422 - ops.cpp:6210-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3421 - ops.cpp:6220-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3431 - ops.cpp:6210-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3402 - ops.cpp:6365-6484 - libggml-cpu.so [...]+0.040.010.010.070.030.010.001890.010.01490.10
Loop 3409 - ops.cpp:6413-6426 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 3410 - ops.cpp:6429-6442 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 3407 - ops.cpp:6446-6457 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 3404 - ops.cpp:6479-6484 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 3411 - ops.cpp:6429-6442 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 3403 - ops.cpp:6462-6475 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 3408 - ops.cpp:6413-6426 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 3406 - ops.cpp:6446-6457 - libggml-cpu.so0.060.000.000.040.040.000.00190.020.0147.16
Loop 3405 - ops.cpp:6479-6484 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 3426 - ops.cpp:6210-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3413 - ops.cpp:6210-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3416 - ops.cpp:6210-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 3418 - ops.cpp:6220-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
kmp_flag_native<unsigned long long, (flag_type)1, true>::done_check()libiomp5.so0.360.020.020.250.250.010.011370.070.04OMP (%): 100.0040.29
__kmp_barrierlibiomp5.so0.440.010.010.300.300.010.011630.040.02OMP (%): 100.00102.92
ggml_compute_forward_add_non_quantized+libggml-cpu.so0.250.010.010.170.030.000.00590.030.02/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.0027.17clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 878 - binary-ops.cpp:10-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 879 - binary-ops.cpp:10-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 880 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 881 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 868 - binary-ops.cpp:10-97 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 870 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 869 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 864 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 865 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 867 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 866 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 890 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 891 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 892 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 895 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 896 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 894 - binary-ops.cpp:10-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 893 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 900 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.010.000.000.160.010.000.0010.000.000.00
Loop 903 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 904 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 905 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 906 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 902 - binary-ops.cpp:10-101 - libggml-cpu.so [...]+0.000.000.000.150.000.000.0000.000.000.00
Loop 901 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.220.000.000.150.150.000.0010.000.00164.71
Loop 882 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 883 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 884 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 885 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 886 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 887 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 889 - binary-ops.cpp:10-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 888 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 897 - binary-ops.cpp:10-97 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 899 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 898 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 875 - binary-ops.cpp:10-97 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 876 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 877 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 871 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 872 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 874 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 873 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 860 - binary-ops.cpp:10-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 863 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 861 - ggml-impl.h:518-541 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 862 - ggml-impl.h:518-541 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
__libm_expf_l9exec0.100.010.010.070.070.000.00330.030.02Math (%): 100.001765.19
ggml_compute_forward_mul+libggml-cpu.so0.120.000.000.080.020.000.00380.020.01/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.0061.41clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 1130 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1132 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1131 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1112 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1113 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1115 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1114 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1126 - binary-ops.cpp:18-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1127 - binary-ops.cpp:18-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1129 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1128 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1108 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1111 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1110 - ggml-impl.h:518-541 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1109 - ggml-impl.h:518-541 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1145 - binary-ops.cpp:18-97 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1147 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1146 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1133 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1137 - binary-ops.cpp:18-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1136 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1134 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1135 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1140 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1143 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1144 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1142 - binary-ops.cpp:18-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1141 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1148 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.080.000.000.0010.000.000.00
Loop 1152 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1154 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1153 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1150 - binary-ops.cpp:18-101 - libggml-cpu.so [...]+0.000.000.000.080.000.000.0000.000.000.00
Loop 1149 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.120.000.000.080.080.000.0010.000.00356.35
Loop 1151 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1119 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1120 - binary-ops.cpp:18-124 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1121 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1122 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1116 - binary-ops.cpp:18-97 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1117 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1118 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1123 - binary-ops.cpp:18-97 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1125 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1124 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1138 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 1139 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
ggml_vec_swiglu_f32+libggml-cpu.so0.350.000.000.240.010.000.0070.150.09/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.003571.89clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 1792 - vec.cpp:402-403 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1791 - vec.cpp:402-403 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 1793 - vec.h:1084-1116 - libggml-cpu.so [...]0.330.000.000.230.230.000.0070.140.093681.88
ggml_compute_forward_rms_norm+libggml-cpu.so0.150.000.000.100.010.000.0060.060.04/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/../icx_3/bin/libggml-blas.so (%): 100.00440.32clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 2854 - ops.cpp:4319-4333 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2855 - ops.cpp:4320-4333 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2856 - vec.h:677-682 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2857 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2853 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2809 - ops.cpp:4319-4355 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2812 - ops.cpp:4320-4355 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2815 - vec.h:687-688 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2816 - vec.h:687-688 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2813 - vec.h:687-688 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2814 - vec.h:687-688 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2810 - vec.h:687-688 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2811 - vec.h:687-688 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2817 - ops.cpp:4319-4355 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2818 - ops.cpp:4320-4355 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2819 - vec.h:687-688 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2821 - ops.cpp:4325-4326 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2822 - ops.cpp:4325-4326 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2820 - vec.h:687-688 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2840 - ops.cpp:4319-4355 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2841 - ops.cpp:4320-4355 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2842 - vec.h:677-688 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2839 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2843 - vec.h:687-688 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2844 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2836 - ops.cpp:4319-4333 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2838 - ops.cpp:4320-4333 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2837 - ops.cpp:4321-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 2835 - ops.cpp:4320-4333 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2834 - ops.cpp:4321-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 2832 - ops.cpp:4321-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 2831 - ops.cpp:4320-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 2833 - ops.cpp:4321-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 2859 - ops.cpp:4319-4338 - libggml-cpu.so [...]+0.000.000.000.100.000.000.0010.000.000.00
Loop 2860 - ops.cpp:4320-4338 - libggml-cpu.so [...]+0.000.000.000.100.000.000.0000.000.000.00
Loop 2861 - vec.h:673-682 - libggml-cpu.so [...]+0.000.000.000.100.000.000.0000.000.000.00
Loop 2864 - ops.cpp:4325-4326 - libggml-cpu.so0.130.000.000.090.090.000.0010.000.00554.33
Loop 2862 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2863 - ops.cpp:4325-4326 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2858 - vec.h:677-682 - libggml-cpu.so0.010.000.000.010.010.000.0010.000.001519.37
Loop 2846 - ops.cpp:4319-4355 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2847 - ops.cpp:4320-4355 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2848 - vec.h:673-688 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2845 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2851 - ops.cpp:4325-4326 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2852 - ops.cpp:4325-4326 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2850 - vec.h:677-682 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2849 - vec.h:687-688 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.000.00
Loop 2828 - ops.cpp:4319-4333 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2827 - ops.cpp:4320-4333 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2825 - ops.cpp:4321-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 2824 - ops.cpp:4321-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 2826 - ops.cpp:4321-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 2823 - ops.cpp:4320-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
Loop 2830 - ops.cpp:4320-4333 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.000.00
Loop 2829 - ops.cpp:4321-4333 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.000.00
_ZN14common_sampler10set_logitsEP13llama_contexti.A+exec0.100.000.000.070.000.000.0010.000.00Exe (%): 100.000.00clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --driver-mode=g++ --intel -D GGML_BACKEND_SHARED -D GGML_SHARED -D GGML_USE_BLAS -D GGML_USE_CPU -D LLAMA_SHARED -...
Loop 3924 - sampling.cpp:125-126 - exec0.100.000.000.070.070.000.0010.000.000.00
Loop 3927 - stl_algobase.h:911-912 - exec0.000.000.000.000.000.000.0000.000.000.00
Loop 3923 - sampling.cpp:125-126 - exec0.000.000.000.000.000.000.0000.000.000.00
Loop 3926 - stl_algobase.h:911-912 - exec0.000.000.000.000.000.000.0000.000.000.00
Loop 3925 - stl_algobase.h:911-912 - exec0.000.000.000.000.000.000.0000.000.000.00
Loop 3928 - stl_algobase.h:911-912 - exec0.000.000.000.000.000.000.0000.000.000.00
×