options

Functions and Loops

Columns Filter

Max Thread Time / Walltime orig_0 (%) Coverage orig_0 (%) Coverage Excluding Loops orig_0 (%) Max Inclusive Time Over Threads orig_0 (s) Max Exclusive Time Over Threads orig_0 (s) Inclusive Time w.r.t. Wall Time orig_0 (s) Exclusive Time w.r.t. Wall Time orig_0 (s) Nb Threads orig_0 Deviation (coverage) orig_0 Deviation (walltime) orig_0 Categories orig_0 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 Compilation Options
NameModuleMax Thread Time / Walltime orig_0 (%)Coverage orig_0 (%)Coverage Excluding Loops orig_0 (%)Max Inclusive Time Over Threads orig_0 (s)Max Exclusive Time Over Threads orig_0 (s)Inclusive Time w.r.t. Wall Time orig_0 (s)Exclusive Time w.r.t. Wall Time orig_0 (s)Nb Threads orig_0Deviation (coverage) orig_0Deviation (walltime) orig_0Categories orig_0Compilation Options
kai_run_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm+libggml-cpu.so34.7353.680.023.370.013.820.00642.230.09/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLE...
Loop 2404 - kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c:131-131 - libggml-cpu.so+0.000.000.000.000.000.000.0000.000.00
Loop 2403 - kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c:131-131 - libggml-cpu.so+0.000.000.000.000.000.000.0000.000.00
Loop 2402 - kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c:131-131 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 2407 - kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c:131-131 - libggml-cpu.so+0.0053.670.003.400.003.820.0000.000.00
Loop 2406 - kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c:131-131 - libggml-cpu.so+0.4153.670.263.400.043.820.02610.140.01
Loop 2405 - kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c:131-131 - libggml-cpu.so34.6253.4053.403.363.363.803.80642.230.09
kmp_flag_64<false, true>::wait(kmp_info*, int, void*)libomp.so15.5917.7417.741.511.511.261.26642.010.13OMP (%): 100.00
ggml_vec_dot_q6_K_q8_K+libggml-cpu.so11.5217.530.091.120.021.250.01641.130.05/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLE...
Loop 2286 - quants.c:2683-2812 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 2289 - quants.c:2492-2660 - libggml-cpu.so [...]+1.8517.432.121.180.181.240.15640.400.02
Loop 2288 - quants.c:2506-2590 - libggml-cpu.so [...]10.2415.3215.321.001.001.091.09641.130.05
Loop 2287 - quants.c:2683-2758 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
kmp_flag_native<unsigned long long, (flag_type)1, true>::notdone_check()libomp.so1.851.931.930.180.180.140.14640.490.03OMP (%): 100.00
$xlibc.so.61.341.161.160.130.130.080.08630.440.03System (%): 100.00
ggml_compute_forward_flash_attn_ext+libggml-cpu.so1.131.140.020.110.010.080.00640.350.02/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 1745 - ops.cpp:8778-8920 - libggml-cpu.so [...]+0.001.120.000.160.000.080.0000.000.00
Loop 1746 - vec.h:282-725 - libggml-cpu.so [...]+0.001.120.000.160.000.080.0000.000.00
Loop 1744 - vec.h:282-725 - libggml-cpu.so [...]+0.211.120.040.160.020.080.00210.060.00
Loop 1760 - vec.h:646-653 - libggml-cpu.so0.100.020.020.010.010.000.00110.030.00
Loop 1750 - ops.cpp:8793-8881 - libggml-cpu.so [...]+0.771.060.730.130.080.080.05630.260.02
Loop 1754 - vec.h:290-338 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 1759 - vec.h:710-717 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 1752 - vec.h:290-338 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 1751 - vec.h:343-348 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 1757 - vec.h:411-458 - libggml-cpu.so0.410.320.320.040.040.020.02610.160.01
Loop 1756 - vec.h:461-466 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 1758 - vec.h:710-717 - libggml-cpu.so0.100.010.010.010.010.000.0070.030.00
Loop 1755 - vec.h:646-653 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 1753 - vec.h:343-348 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 1749 - ops.cpp:8885-8886 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 1748 - ops.cpp:8885-8886 - libggml-cpu.so [...]0.050.000.000.010.010.000.0020.000.00
Loop 1747 - vec.h:646-653 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
ggml_compute_forward_rope_f32(ggml_compute_params const*, ggml_tensor*, bool)+libggml-cpu.so0.820.830.010.080.010.060.00640.240.01/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 1434 - ops.cpp:6210-6484 - libggml-cpu.so [...]+0.100.820.020.110.010.060.00160.020.00
Loop 1433 - ops.cpp:6210-6409 - libggml-cpu.so [...]+0.150.720.060.080.020.050.00350.050.00
Loop 1440 - ops.cpp:6210-6303 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 1443 - ops.cpp:6210-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 1441 - ops.cpp:6220-6245 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 1442 - ops.cpp:6220-6245 - libggml-cpu.so [...]0.670.660.660.070.070.050.05640.230.01
Loop 1439 - ops.cpp:6429-6442 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 1435 - ops.cpp:6462-6475 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 1438 - ops.cpp:6413-6426 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 1437 - ops.cpp:6446-6456 - libggml-cpu.so [...]0.210.080.080.020.020.010.01380.070.00
Loop 1436 - ops.cpp:6479-6484 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
__pthread_mutex_locklibc.so.60.980.690.690.090.090.050.05610.410.02Pthread (%): 100.00
__sched_yieldlibc.so.60.770.670.670.070.070.050.05630.230.01OMP (%): 100.00
ggml_vec_swiglu_f32+libggml-cpu.so2.060.540.000.200.000.040.00160.530.03/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 908 - vec.cpp:402-403 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 909 - vec.cpp:403-403 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 910 - vec.cpp:385-387 - libggml-cpu.so [...]2.060.540.540.200.200.040.04160.530.03
ggml_vec_dot_f16+libggml-cpu.so0.620.480.040.060.010.030.00640.180.01/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 902 - vec.cpp:266-269 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 903 - vec.cpp:231-262 - libggml-cpu.so0.570.440.440.050.050.030.03640.190.01
unknown_function[vdso]0.460.320.000.050.000.020.00630.170.01OMP (%): 100.00
__aarch64_ldadd8_acq_rellibomp.so0.980.260.260.090.090.020.02560.250.02OMP (%): 100.00
__expf_finitelibamath.so0.360.250.250.040.040.020.02620.130.01Math (%): 100.00
__sincosf_finitelibamath.so0.410.230.230.040.040.020.02610.140.01Math (%): 100.00
kai_run_lhs_quant_pack_qsi8d32p4x8sb_f32_neon+libggml-cpu.so2.470.230.000.240.000.020.0040.330.02/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLE...
Loop 2366 - kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c:268-335 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 2367 - kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c:271-332 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 2369 - kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c:93-264 - libggml-cpu.so [...]+0.000.230.000.240.000.020.0000.000.00
Loop 2368 - kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c:96-258 - libggml-cpu.so [...]2.470.230.230.240.240.020.0240.330.02
ggml_graph_compute_thread+libggml-cpu.so0.360.220.040.040.020.020.00550.140.01/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLE...
Loop 87 - ggml-cpu.c:1592-1601 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 88 - ggml-cpu.c:1436-1642 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 96 - ggml-cpu.c:1436-1465 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 95 - ggml-cpu.c:1436-1465 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 94 - ggml-cpu.c:1438-1465 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 93 - ggml-cpu.c:1461-1462 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 92 - ggml-cpu.c:1436-1465 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 91 - ggml-cpu.c:1436-1465 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 90 - ggml-cpu.c:1438-1465 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 89 - ggml-cpu.c:1454-1462 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 97 - ggml-cpu.c:1585-1587 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 86 - ggml-cpu.c:533-2897 - libggml-cpu.so [...]+0.360.180.180.030.030.010.01520.130.01
Loop 85 - ggml-cpu.c:533-2897 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 99 - ggml-cpu.c:1572-1579 - libggml-cpu.so+0.000.000.000.000.000.000.0000.000.00
Loop 98 - ggml-cpu.c:1573-1579 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 102 - ggml-cpu.c:1552-1560 - libggml-cpu.so+0.000.000.000.000.000.000.0000.000.00
Loop 101 - ggml-cpu.c:1552-1560 - libggml-cpu.so+0.000.000.000.000.000.000.0000.000.00
Loop 100 - ggml-cpu.c:1552-1560 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 103 - ggml-cpu.c:2087-2088 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
__kmp_hyper_barrier_release(barrier_type, kmp_info*, int, int, int, void*)libomp.so0.360.210.210.040.040.010.01590.120.01OMP (%): 100.00
kai_run_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0+libggml-cpu.so7.000.180.000.680.000.010.0010.000.00/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLE...
Loop 2380 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:107-154 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 2382 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:107-154 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 2383 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:115-118 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 2381 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:127-142 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 2384 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:107-154 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 2386 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:115-118 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 2385 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:127-134 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 2378 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:107-154 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 2379 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:107-154 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 2377 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:123-154 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 2376 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:127-148 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 2375 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:145-148 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 2374 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:115-118 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 2387 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:107-154 - libggml-cpu.so [...]+1.030.180.030.680.100.010.0010.000.00
Loop 2388 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:127-139 - libggml-cpu.so [...]4.170.110.110.410.410.010.0110.000.00
Loop 2389 - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:115-118 - libggml-cpu.so1.800.050.050.170.170.000.0010.000.00
ggml_compute_forward_mul_mat+libggml-cpu.so0.210.140.000.020.000.010.00510.080.00/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLE...
Loop 66 - ggml-cpu.c:1125-1395 - libggml-cpu.so [...]+0.050.140.000.070.000.010.0010.000.00
Loop 67 - ggml-cpu.c:1162-1198 - libggml-cpu.so [...]+0.050.100.000.050.000.010.0010.000.00
Loop 68 - ggml-cpu.c:1162-1198 - libggml-cpu.so [...]+0.050.100.010.050.010.010.0040.010.00
Loop 69 - ggml-cpu.c:1163-1198 - libggml-cpu.so [...]+0.150.100.020.040.010.010.00120.050.00
Loop 65 - ggml-cpu.c:1183-1194 - libggml-cpu.so [...]0.150.080.080.020.020.010.01410.050.00
Loop 70 - ggml-cpu.c:1197-1198 - libggml-cpu.so0.100.000.000.010.010.000.0020.050.00
Loop 64 - ggml-cpu.c:1125-1395 - libggml-cpu.so [...]0.100.030.030.010.010.000.00240.020.00
Loop 73 - ggml-cpu.c:1289-1297 - libggml-cpu.so+0.000.000.000.010.000.000.0000.000.00
Loop 72 - ggml-cpu.c:1289-1297 - libggml-cpu.so+0.000.000.000.010.000.000.0000.000.00
Loop 71 - ggml-cpu.c:1289-1297 - libggml-cpu.so0.050.000.000.010.010.000.0010.000.00
ggml_compute_forward_rms_norm+libggml-cpu.so0.620.140.000.060.000.010.00170.320.02/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 1281 - ops.cpp:4319-4365 - libggml-cpu.so [...]+0.000.140.000.070.000.010.0000.000.00
Loop 1283 - ops.cpp:4319-4365 - libggml-cpu.so [...]+0.050.140.000.070.000.010.0020.000.00
Loop 1282 - ops.cpp:4325-4326 - libggml-cpu.so0.570.120.120.050.050.010.01160.280.02
Loop 1284 - ops.cpp:4325-4326 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 1285 - vec.h:646-653 - libggml-cpu.so0.150.010.010.010.010.000.0070.070.00
ggml_compute_forward_add_non_quantized+libggml-cpu.so0.510.120.010.050.010.010.00230.230.01/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 430 - binary-ops.cpp:10-110 - libggml-cpu.so [...]+0.050.110.000.060.000.010.0010.000.00
Loop 432 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 431 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.510.110.110.050.050.010.01160.200.01
Loop 429 - binary-ops.cpp:84-84 - libggml-cpu.so0.050.000.000.010.010.000.0010.000.00
Loop 433 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 411 - binary-ops.cpp:10-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 406 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 408 - binary-ops.cpp:10-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 410 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 407 - binary-ops.cpp:84-84 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 409 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 397 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 396 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 398 - binary-ops.cpp:10-95 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 418 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 417 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 419 - binary-ops.cpp:10-95 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 427 - binary-ops.cpp:10-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 428 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 426 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 425 - binary-ops.cpp:84-101 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 393 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 395 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 394 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 392 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 443 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 442 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 413 - binary-ops.cpp:10-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 415 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 414 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 412 - binary-ops.cpp:84-84 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 416 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 422 - binary-ops.cpp:10-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 424 - binary-ops.cpp:10-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 423 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 421 - binary-ops.cpp:84-101 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 420 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 400 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 399 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 402 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 401 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 445 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 444 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 434 - binary-ops.cpp:10-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 436 - binary-ops.cpp:10-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 435 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 437 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 404 - binary-ops.cpp:10-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 403 - binary-ops.cpp:10-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 405 - binary-ops.cpp:10-95 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 438 - binary-ops.cpp:10-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 440 - binary-ops.cpp:10-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 439 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 441 - binary-ops.cpp:10-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
ggml::cpu::kleidiai::extra_buffer_type::get_tensor_traits(ggml_tensor const*)libggml-cpu.so0.260.120.120.030.030.010.01470.090.01/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
ggml_compute_forward_mul+libggml-cpu.so0.360.110.030.040.020.010.00320.140.01/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 532 - binary-ops.cpp:18-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 534 - binary-ops.cpp:18-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 533 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 531 - binary-ops.cpp:84-101 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 530 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 521 - binary-ops.cpp:18-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 516 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 518 - binary-ops.cpp:18-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 519 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 520 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 517 - binary-ops.cpp:84-84 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 554 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 553 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 523 - binary-ops.cpp:18-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 525 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 522 - binary-ops.cpp:84-84 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 526 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 524 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 550 - binary-ops.cpp:18-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 548 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 552 - binary-ops.cpp:18-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 549 - binary-ops.cpp:84-84 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 551 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 540 - binary-ops.cpp:18-110 - libggml-cpu.so [...]+0.000.070.000.040.000.010.0000.000.00
Loop 539 - binary-ops.cpp:84-84 - libggml-cpu.so0.000.000.000.000.000.000.0000.000.00
Loop 543 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 542 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 541 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.360.070.070.040.040.010.01150.130.01
Loop 556 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 555 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 544 - binary-ops.cpp:18-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 546 - binary-ops.cpp:18-110 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 545 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 547 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 528 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 527 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 529 - binary-ops.cpp:18-95 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 537 - binary-ops.cpp:18-101 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 535 - binary-ops.cpp:84-101 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 538 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 536 - binary-ops.cpp:18-32 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 507 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 508 - binary-ops.cpp:18-95 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 506 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 510 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 509 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 512 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 511 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 514 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 513 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 515 - binary-ops.cpp:18-95 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 503 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 502 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
Loop 505 - binary-ops.cpp:18-95 - libggml-cpu.so [...]+0.000.000.000.000.000.000.0000.000.00
Loop 504 - binary-ops.cpp:18-45 - libggml-cpu.so [...]0.000.000.000.000.000.000.0000.000.00
__memcpylibastring.so0.770.090.090.080.080.010.01330.170.01String (%): 100.00
__kmp_hyper_barrier_gather(barrier_type, kmp_info*, int, int, void (*)(void*, void*), void*)libomp.so0.260.090.090.030.030.010.01370.090.00OMP (%): 100.00
ggml_cpu_fp32_to_fp16+libggml-cpu.so0.260.090.000.020.000.010.00390.080.01/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLE...
Loop 0 - ggml-cpu.c:3228-3229 - libggml-cpu.so [...]0.260.090.090.020.020.010.01380.080.01
kmp_flag_native<unsigned long long, (flag_type)1, true>::done_check()libomp.so0.410.080.080.040.040.010.01190.140.01OMP (%): 100.00
unknown_functionlibggml-cpu.so0.210.070.000.020.000.000.00370.060.00/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00
__kmp_barrierlibomp.so0.310.070.070.030.030.000.00320.080.01OMP (%): 100.00
__GI___lll_lock_waitlibc.so.60.150.050.050.020.020.000.00270.050.00Pthread (%): 100.00
@plt_start@libomp.so0.210.050.050.020.020.000.00260.060.00OMP (%): 100.00
__GI___lll_lock_wakelibc.so.60.150.050.050.010.010.000.00260.050.00Pthread (%): 94.44
System (%): 5.56
$xlibc.so.60.150.040.040.020.020.000.00220.060.00System (%): 100.00
$xlibc.so.60.150.040.040.010.010.000.00230.050.00Pthread (%): 100.00
__kmpc_barrierlibomp.so0.100.030.030.010.010.000.00190.030.00OMP (%): 100.00
ggml_compute_forward_set_rows+libggml-cpu.so0.100.020.020.010.010.000.00160.030.00/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 1364 - ops.cpp:5550-5563 - libggml-cpu.so+0.000.000.000.010.000.000.0000.000.00
Loop 1363 - ops.cpp:5551-5563 - libggml-cpu.so+0.000.000.000.010.000.000.0000.000.00
Loop 1362 - ops.cpp:5552-5563 - libggml-cpu.so0.050.000.000.010.010.000.0010.000.00
__fs_pow_1libamath.so0.050.020.020.010.010.000.00170.000.00Math (%): 100.00
ggml_is_emptylibggml-base.so0.100.020.020.010.010.000.00140.030.00/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml-blas.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 -D GGML_BUILD -D GGML_COMMIT="unknown" -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLEIDIAI ...
ggml::cpu::kleidiai::tensor_traits::compute_forward_q4_0(ggml_compute_params*, ggml_tensor*)libggml-cpu.so0.050.020.020.010.010.000.00140.000.00/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
__memsetlibastring.so0.460.020.020.050.050.000.0040.260.02String (%): 100.00
__kmp_now_nseclibomp.so0.100.020.020.010.010.000.00120.020.00OMP (%): 100.00
ggml::cpu::repack::extra_buffer_type::get_tensor_traits(ggml_tensor const*)libggml-cpu.so0.100.010.010.010.010.000.00100.030.00/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
__kmp_yieldlibomp.so0.100.010.010.010.010.000.0060.050.00OMP (%): 100.00
ggml_cpu_extra_compute_forward+libggml-cpu.so0.050.010.000.010.010.000.0090.000.00/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/176-138-1719/llama.cpp/build/llama.cpp/../build/bin/libggml.so (%): 100.00Arm C/C++/Fortran Compiler version 24.10.1 (build number 4) (based on LLVM 19.1.0) /opt/arm/arm-linux-compiler-24.10.1_Ubuntu-20.04/llvm-bin/clang-19 --driver-mode=g++ -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -...
Loop 390 - traits.cpp:13-17 - libggml-cpu.so [...]0.050.010.010.010.010.000.0060.000.00
__ieee754_log2libamath.so0.050.010.010.010.010.000.0080.000.00Math (%): 100.00
__exp2f_finitelibamath.so0.100.010.010.010.010.000.0070.030.00Math (%): 100.00
×