Help is available by moving the cursor above any
symbol or by checking MAQAO website.
| Metric | r0 | r1 | r2 | r3 | r4 | r5 | r6 | r7 | r8 | r9 | r10 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total Time (s) | 184.23 | 76.07 | 39.51 | 21.11 | 12.00 | 9.66 | 8.26 | 7.41 | 6.78 | 6.31 | 5.90 | |
| Max (Thread Active Time) (s) | 147.73 | 74.31 | 37.82 | 19.36 | 10.13 | 8.01 | 6.47 | 5.75 | 5.02 | 4.53 | 4.20 | |
| Average Active Time (s) | 147.73 | 73.92 | 37.21 | 18.71 | 9.52 | 7.17 | 5.75 | 4.87 | 4.23 | 3.75 | 3.32 | |
| Activity Ratio (%) | 80.2 | 98.6 | 98.3 | 97.6 | 96.1 | 95.1 | 94.1 | 93.0 | 91.9 | 91.1 | 89.7 | |
| Average number of active threads | 0.802 | 1.944 | 3.767 | 7.091 | 12.699 | 17.821 | 22.269 | 26.267 | 29.998 | 33.275 | 36.033 | |
| Affinity Stability (%) | 71.2 | 98.4 | 98.1 | 97.5 | 96.3 | 95.8 | 95.0 | 94.4 | 93.8 | 93.2 | 92.5 | |
| Time in analyzed loops (%) | 99.1 | 99.0 | 98.6 | 98.0 | 96.9 | 95.2 | 94.9 | 93.0 | 91.8 | 90.7 | 91.1 | |
| Time in analyzed innermost loops (%) | 96.7 | 96.6 | 96.3 | 95.7 | 94.5 | 93.1 | 92.6 | 91.0 | 90.0 | 88.8 | 88.7 | |
| Time in user code (%) | 99.3 | 99.2 | 98.9 | 98.2 | 97.1 | 95.5 | 95.1 | 93.2 | 92.0 | 91.0 | 91.3 | |
| Compilation Options Score (%) | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 99.9 | 99.9 | 99.9 | 99.9 | 99.9 | 99.9 | |
| Array Access Efficiency (%) | 74.1 | 74.1 | 74.1 | 74.1 | 74.2 | 74.2 | 74.2 | 74.2 | 74.2 | 74.3 | 74.1 | |
| Potential Speedups | ||||||||||||
| Perfect Flow Complexity | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | |
| Perfect OpenMP/MPI/Pthread/TBB | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.01 | 1.01 | 1.01 | 1.01 | 1.02 | 1.02 | |
| Perfect OpenMP/MPI/Pthread/TBB + Perfect Load Distribution | 1.00 | 1.01 | 1.02 | 1.05 | 1.09 | 1.16 | 1.17 | 1.25 | 1.28 | 1.32 | 1.37 | |
| Scalability - Gap | 1.00 | 0.83 | 0.86 | 0.92 | 1.04 | 1.26 | 1.43 | 1.61 | 1.77 | 1.92 | 2.05 | |
| No Scalar Integer | Potential Speedup | 1.01 | 1.01 | 1.01 | 1.01 | 1.01 | 1.01 | 1.01 | 1.01 | 1.01 | 1.01 | 1.01 |
| Nb Loops to get 80% | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | |
| FP Vectorised | Potential Speedup | 1.17 | 1.17 | 1.17 | 1.17 | 1.17 | 1.16 | 1.16 | 1.16 | 1.16 | 1.16 | 1.16 |
| Nb Loops to get 80% | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | |
| Fully Vectorised | Potential Speedup | 4.00 | 3.99 | 3.93 | 3.88 | 3.76 | 3.63 | 3.57 | 3.41 | 3.31 | 3.23 | 3.22 |
| Nb Loops to get 80% | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | |
| Only FP Arithmetic | Potential Speedup | 4.58 | 4.57 | 4.49 | 4.38 | 4.20 | 3.97 | 3.91 | 3.66 | 3.53 | 3.42 | 3.43 |
| Nb Loops to get 80% | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | |
| Source Object | Issue |
|---|---|
| ▼libllama.so | |
| ▼hashtable.h | |
| ○ | -mcpu=native is missing. |
| ▼llama-vocab.cpp | |
| ○ | -mcpu=native is missing. |
| ▼llama-impl.cpp | |
| ○ | -mcpu=native is missing. |
| ▼hashtable_policy.h | |
| ○ | -mcpu=native is missing. |
| ▼libggml-cpu.so | |
| ▼vec.cpp | |
| ○ | |
| ▼traits.cpp | |
| ○ | |
| ▼kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c | |
| ○ | |
| ▼kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c | |
| ○ | |
| ▼kleidiai.cpp | |
| ○ | |
| ▼ops.cpp | |
| ○ | |
| ▼kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c | |
| ○ | |
| ▼binary-ops.cpp | |
| ○ | |
| ▼ggml-cpu.c | |
| ○ | |
| ▼quants.c | |
| ○ | |
| ▼libggml-base.so | |
| ▼ | |
| ○ | -g is missing for some functions (possibly ones added by the compiler), it is needed to have more accurate reports. Other recommended flags are: -O2/-O3, -march=(target) |
| ○ | -O2, -O3 or -Ofast is missing. |
| ○ | -mcpu=native is missing. |
| ▼[vdso] | |
| ▼ | |
| ○ | -g is missing for some functions (possibly ones added by the compiler), it is needed to have more accurate reports. Other recommended flags are: -O2/-O3, -march=(target) |
| ○ | -O2, -O3 or -Ofast is missing. |
| ○ | -mcpu=native is missing. |
| ▼exec | |
| ▼ | |
| ○ | -g is missing for some functions (possibly ones added by the compiler), it is needed to have more accurate reports. Other recommended flags are: -O2/-O3, -march=(target) |
| ○ | -O2, -O3 or -Ofast is missing. |
| ○ | -mcpu=native is missing. |
| r0 | r1 | r2 | r3 | r4 | r5 | r6 | r7 | r8 | r9 | r10 | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Application | /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/run/oneview_runs/defaults/orig/exec | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Timestamp | 2025-11-24 12:08:25 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Experiment Type | MPI; | MPI; OpenMP; | same as r1 | same as r1 | same as r1 | same as r1 | same as r1 | same as r1 | same as r1 | same as r1 | same as r1 |
| Machine | ip-172-31-46-37.ec2.internal | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Architecture | aarch64 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Micro Architecture | ARM_NEOVERSE_V1 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Model Name | |||||||||||
| Cache Size | |||||||||||
| Number of Cores | |||||||||||
| Maximal Frequency | 0 GHz | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| OS Version | Linux 6.1.158-178.288.amzn2023.aarch64 #1 SMP Mon Nov 3 18:38:05 UTC 2025 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Architecture used during static analysis | aarch64 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Micro Architecture used during static analysis | ARM_NEOVERSE_V1 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Compilation Options | libggml-base.so: N/A libggml-cpu.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLEIDIAI -D GGML_USE_CPU_REPACK -D GGML_USE_LLAMAFILE -D GGML_USE_OPENMP -D _GNU_SOURCE -D _XOPEN_SOURCE=600 -D ggml_cpu_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/.. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_fp32_bf16p_bf16p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -mcpu=native+dotprod+i8mm+sve+nosme -fopenmp=libomp -mcpu=native+dotprod+i8mm+sve+nosme -MD -MT ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -MF ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o.d -o ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c libllama.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 --driver-mode=g++ -D GGML_BACKEND_SHARED -D GGML_SHARED -D GGML_USE_BLAS -D GGML_USE_CPU -D LLAMA_BUILD -D LLAMA_SHARED -D llama_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/../include -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -MD -MT src/CMakeFiles/llama.dir/unicode.cpp.o -MF src/CMakeFiles/llama.dir/unicode.cpp.o.d -o src/CMakeFiles/llama.dir/unicode.cpp.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/unicode.cpp | same as r0 | + [vdso]: N/A exec: N/A libggml-base.so: N/A libggml-cpu.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLEIDIAI -D GGML_USE_CPU_REPACK -D GGML_USE_LLAMAFILE -D GGML_USE_OPENMP -D _GNU_SOURCE -D _XOPEN_SOURCE=600 -D ggml_cpu_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/.. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_fp32_bf16p_bf16p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -mcpu=native+dotprod+i8mm+sve+nosme -fopenmp=libomp -mcpu=native+dotprod+i8mm+sve+nosme -MD -MT ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -MF ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o.d -o ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c libllama.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 --driver-mode=g++ -D GGML_BACKEND_SHARED -D GGML_SHARED -D GGML_USE_BLAS -D GGML_USE_CPU -D LLAMA_BUILD -D LLAMA_SHARED -D llama_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/../include -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -MD -MT src/CMakeFiles/llama.dir/unicode.cpp.o -MF src/CMakeFiles/llama.dir/unicode.cpp.o.d -o src/CMakeFiles/llama.dir/unicode.cpp.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/unicode.cpp | same as r2 | + [vdso]: N/A libggml-cpu.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLEIDIAI -D GGML_USE_CPU_REPACK -D GGML_USE_LLAMAFILE -D GGML_USE_OPENMP -D _GNU_SOURCE -D _XOPEN_SOURCE=600 -D ggml_cpu_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/.. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_fp32_bf16p_bf16p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -mcpu=native+dotprod+i8mm+sve+nosme -fopenmp=libomp -mcpu=native+dotprod+i8mm+sve+nosme -MD -MT ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -MF ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o.d -o ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c libllama.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 --driver-mode=g++ -D GGML_BACKEND_SHARED -D GGML_SHARED -D GGML_USE_BLAS -D GGML_USE_CPU -D LLAMA_BUILD -D LLAMA_SHARED -D llama_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/../include -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -MD -MT src/CMakeFiles/llama.dir/unicode.cpp.o -MF src/CMakeFiles/llama.dir/unicode.cpp.o.d -o src/CMakeFiles/llama.dir/unicode.cpp.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/unicode.cpp | + [vdso]: N/A exec: N/A libggml-cpu.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLEIDIAI -D GGML_USE_CPU_REPACK -D GGML_USE_LLAMAFILE -D GGML_USE_OPENMP -D _GNU_SOURCE -D _XOPEN_SOURCE=600 -D ggml_cpu_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/.. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_fp32_bf16p_bf16p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -mcpu=native+dotprod+i8mm+sve+nosme -fopenmp=libomp -mcpu=native+dotprod+i8mm+sve+nosme -MD -MT ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -MF ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o.d -o ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c libllama.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 --driver-mode=g++ -D GGML_BACKEND_SHARED -D GGML_SHARED -D GGML_USE_BLAS -D GGML_USE_CPU -D LLAMA_BUILD -D LLAMA_SHARED -D llama_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/../include -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -MD -MT src/CMakeFiles/llama.dir/unicode.cpp.o -MF src/CMakeFiles/llama.dir/unicode.cpp.o.d -o src/CMakeFiles/llama.dir/unicode.cpp.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/unicode.cpp | + [vdso]: N/A libggml-base.so: N/A libggml-cpu.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_KLEIDIAI -D GGML_USE_CPU_REPACK -D GGML_USE_LLAMAFILE -D GGML_USE_OPENMP -D _GNU_SOURCE -D _XOPEN_SOURCE=600 -D ggml_cpu_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/.. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_fp32_bf16p_bf16p -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -mcpu=native+dotprod+i8mm+sve+nosme -fopenmp=libomp -mcpu=native+dotprod+i8mm+sve+nosme -MD -MT ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -MF ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o.d -o ggml/src/CMakeFiles/ggml-cpu.dir/__/__/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c libllama.so: Arm Toolchain for Linux 21.1 clang version 21.1.1 (https://github.com/arm/arm-toolchain.git 8406e267866a1511ab7f7d56ef7361208c0d44ae) /opt/arm/arm-toolchain-for-linux/bin/clang-21 -frtlib-add-rpath -fveclib=ArmPL -mllvm -gvn-add-phi-translation=1 -mllvm -store-to-load-forwarding-conflict-detection=0 --driver-mode=g++ -D GGML_BACKEND_SHARED -D GGML_SHARED -D GGML_USE_BLAS -D GGML_USE_CPU -D LLAMA_BUILD -D LLAMA_SHARED -D llama_EXPORTS -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/. -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/../include -I /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -O3 -D NDEBUG -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -MD -MT src/CMakeFiles/llama.dir/unicode.cpp.o -MF src/CMakeFiles/llama.dir/unicode.cpp.o.d -o src/CMakeFiles/llama.dir/unicode.cpp.o -c /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-46-37.ec2.internal/176-398-0068/llama.cpp/build/llama.cpp/src/unicode.cpp | same as r6 | same as r6 | same as r6 | same as r2 |
| Number of processes observed | 1 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Number of threads observed | 1 | 2 | 4 | 8 | 16 | 24 | 32 | 40 | 48 | 56 | 64 |
| Frequency Driver | NA | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Frequency Governor | NA | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Huge Pages | madvise | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Hyperthreading | off | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Number of sockets | 1 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Number of cores per socket | 64 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| MAQAO version | 2025.1.3 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| MAQAO build | b489783858807c9a72e0923fd7b399a22c81991c::20251024-122946 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| Comments | OV scalability run using armclang | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |