==2626== Cachegrind, a cache and branch-prediction profiler
==2626== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2626== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2626== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/t9300-fast-import.sh
==2626== Parent PID: 2619
==2626== 
--2626-- warning: L3 cache found, using its data for the LL simulation.
==2626== 
==2626== I   refs:      13,977,140
==2626== I1  misses:        38,057
==2626== LLi misses:         8,097
==2626== I1  miss rate:       0.27%
==2626== LLi miss rate:       0.06%
==2626== 
==2626== D   refs:       6,295,500  (3,526,460 rd   + 2,769,040 wr)
==2626== D1  misses:        64,598  (   31,439 rd   +    33,159 wr)
==2626== LLd misses:        43,550  (   13,264 rd   +    30,286 wr)
==2626== D1  miss rate:        1.0% (      0.9%     +       1.2%  )
==2626== LLd miss rate:        0.7% (      0.4%     +       1.1%  )
==2626== 
==2626== LL refs:          102,655  (   69,496 rd   +    33,159 wr)
==2626== LL misses:         51,647  (   21,361 rd   +    30,286 wr)
==2626== LL miss rate:         0.3% (      0.1%     +       1.1%  )