==2150== Cachegrind, a cache and branch-prediction profiler
==2150== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2150== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2150== Command: bash -n benchmarks/testdata/t9300-fast-import.sh
==2150== Parent PID: 2143
==2150== 
--2150-- warning: L3 cache found, using its data for the LL simulation.
==2150== 
==2150== I   refs:      41,138,083
==2150== I1  misses:         5,049
==2150== LLi misses:         2,593
==2150== I1  miss rate:       0.01%
==2150== LLi miss rate:       0.01%
==2150== 
==2150== D   refs:      17,161,510  (10,763,636 rd   + 6,397,874 wr)
==2150== D1  misses:        12,914  (     9,165 rd   +     3,749 wr)
==2150== LLd misses:         7,947  (     4,836 rd   +     3,111 wr)
==2150== D1  miss rate:        0.1% (       0.1%     +       0.1%  )
==2150== LLd miss rate:        0.0% (       0.0%     +       0.0%  )
==2150== 
==2150== LL refs:           17,963  (    14,214 rd   +     3,749 wr)
==2150== LL misses:         10,540  (     7,429 rd   +     3,111 wr)
==2150== LL miss rate:         0.0% (       0.0%     +       0.0%  )