options

Loops Index

70 loops have been discarded from the report because their coverage is lower than the threshold set by object_coverage_threshold (0.01%). It represents about 1.46% 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 Exclusive Coverage gcc_0 (%) Inclusive Coverage gcc_0 (%) Max Exclusive Time Over Threads gcc_0 (s) Max Inclusive Time Over Threads gcc_0 (s) Exclusive Time w.r.t. Wall Time gcc_0 (s) Inclusive Time w.r.t. Wall Time gcc_0 (s) Nb Threads gcc_0 GFLOPS gcc_0 Vectorization Ratio (%) Vector Length Use (%) Speedup If No Scalar Integer Speedup If FP Vectorized Speedup If Fully Vectorized Speedup If Perfect Load Balancing gcc_0 Stride 0 Stride 1 Stride n Stride Unknown Stride Indirect Array Access Efficiency
Loop idSource LocationSource FunctionLevelExclusive Coverage gcc_0 (%)Inclusive Coverage gcc_0 (%)Max Exclusive Time Over Threads gcc_0 (s)Max Inclusive Time Over Threads gcc_0 (s)Exclusive Time w.r.t. Wall Time gcc_0 (s)Inclusive Time w.r.t. Wall Time gcc_0 (s)Nb Threads gcc_0GFLOPS gcc_0Vectorization Ratio (%)Vector Length Use (%)Speedup If No Scalar IntegerSpeedup If FP VectorizedSpeedup If Fully VectorizedSpeedup If Perfect Load Balancing gcc_0Stride 0Stride 1Stride nStride UnknownStride IndirectArray Access Efficiency
405libggml-cpu.so - mmq.cpp:1597-1597 [...](anonymous namespace)::tinygemm_kernel_vnni<block_q8_0, block_q8_0, float, 1, 64, 32>::apply(int, void const*, void const*, float*, int)Single86.3586.3524.6524.6524.1424.1419289.2587.6785.7111.071.040420178.57
2122libggml-cpu.so - quants.c:298-321 [...]quantize_row_q8_0Single0.030.030.360.360.010.0161070.906028.7511.42.841.48NANANANANA0.00
×