options

Loops Index

27 loops 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 0.03% 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

Columns Filter

Level Max Thread Time / Walltime icx_10 (%) Exclusive Coverage icx_10 (%) Inclusive Coverage icx_10 (%) Max Exclusive Time Over Threads icx_10 (s) Max Inclusive Time Over Threads icx_10 (s) Exclusive Time w.r.t. Wall Time icx_10 (s) Inclusive Time w.r.t. Wall Time icx_10 (s) Nb Threads icx_10 GFLOPS icx_10 Vectorization Ratio (%) Vector Length Use (%) Speedup If No Scalar Integer Speedup If FP Vectorized Speedup If Fully Vectorized Speedup If Perfect Load Balancing icx_10 Stride 0 Stride 1 Stride n Stride Unknown Stride Indirect Array Access Efficiency Level Max Thread Time / Walltime Exclusive Coverage Inclusive Coverage Max Exclusive Time Over Threads Max Inclusive Time Over Threads Exclusive Time w.r.t. Wall Time Inclusive Time w.r.t. Wall Time Nb Threads GFLOPS Vectorization Ratio Vector Length Use Speedup If No Scalar Integer Speedup If FP Vectorized Speedup If Fully Vectorized Speedup If Perfect Load Balancing Stride 0 Stride 1 Stride n Stride Unknown Stride Indirect Array Access Efficiency
Loop idSource LocationSource FunctionLevelMax Thread Time / Walltime icx_10 (%)Exclusive Coverage icx_10 (%)Inclusive Coverage icx_10 (%)Max Exclusive Time Over Threads icx_10 (s)Max Inclusive Time Over Threads icx_10 (s)Exclusive Time w.r.t. Wall Time icx_10 (s)Inclusive Time w.r.t. Wall Time icx_10 (s)Nb Threads icx_10GFLOPS icx_10Vectorization Ratio (%)Vector Length Use (%)Speedup If No Scalar IntegerSpeedup If FP VectorizedSpeedup If Fully VectorizedSpeedup If Perfect Load Balancing icx_10Stride 0Stride 1Stride nStride UnknownStride IndirectArray Access Efficiency
2394libggml-cpu.so - sgemm.cpp:144-399 [...]void (anonymous namespace)::tinyBLAS<16, float __vector(16), float __vector(16), unsigned short, unsigned short, float>::gemm<4, 6, 2>(long, long, long)Innermost0.370.480.480.120.120.070.071923214.07NANANANANA1.74NANANANANA0.00
1644libggml-cpu.so - ops.cpp:5552-5563ggml_compute_forward_set_rowsInnermost0.290.300.300.090.090.040.041924.11012.51182.151004060.00
540libggml-cpu.so - mmq.cpp:303-1392 [...]void parallel_for<(anonymous namespace)::convert_B_packed_format<block_q8_0, 32>(void*, block_q8_0 const*, int, int)::{lambda(int, int)#1}>(int, (anonymous namespace)::convert_B_packed_format<block_q8_0, 32>(void*, block_q8_0 const*, int,...Innermost0.190.120.120.060.060.020.021780.0010041.391.4211.343.0915005087.50
2393libggml-cpu.so - sgemm.cpp:144-464 [...]void (anonymous namespace)::tinyBLAS<16, float __vector(16), float __vector(16), unsigned short, unsigned short, float>::gemm<4, 6, 2>(long, long, long)InBetween0.160.120.600.050.150.020.091923552.9079.3429.861.011.532.72.79NANANANANA50.00
595exec - __intel_avx_rep_memcpySingle0.160.120.120.050.050.020.021870.90100501122.7802000100.00
1724libggml-cpu.so - ops.cpp:6220-6245 [...]ggml_compute_forward_rope_f32(ggml_compute_params const*, ggml_tensor*, bool)Innermost0.140.100.100.040.040.020.02192629.742.176.6611.85.333.0311000100.00
×