options

OpenMP Regions

requested parallelismwalltime sum (s)nb instancesany sync average per thread time (s)any wait average per thread time (s)parallelism overhead (%)local speedup if perfectly balancedglobal speedup if perfectly balanced
start addrfunction namesource locationlevelancestor thread numinvokerparallel or teamsaocc_0aocc_0aocc_0aocc_0aocc_0aocc_0aocc_0aocc_0
libggml-cpu.so:0x2e475ggml_graph_computeggml-cpu.c:314800runtimeparallel19256.40651321.83121.75938.71.6311.625
×