==2581== Cachegrind, a cache and branch-prediction profiler
==2581== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2581== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2581== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/ltmain.sh
==2581== Parent PID: 2574
==2581== 
--2581-- warning: L3 cache found, using its data for the LL simulation.
==2581== brk segment overflow in thread #1: can't grow to 0x482c000
==2581== (see section Limitations in user manual)
==2581== NOTE: further instances of this message will not be shown
==2581== 
==2581== I   refs:      112,762,559
==2581== I1  misses:      1,046,529
==2581== LLi misses:          8,750
==2581== I1  miss rate:        0.93%
==2581== LLi miss rate:        0.01%
==2581== 
==2581== D   refs:       53,193,685  (27,641,342 rd   + 25,552,343 wr)
==2581== D1  misses:        671,572  (   390,540 rd   +    281,032 wr)
==2581== LLd misses:        157,939  (    14,284 rd   +    143,655 wr)
==2581== D1  miss rate:         1.3% (       1.4%     +        1.1%  )
==2581== LLd miss rate:         0.3% (       0.1%     +        0.6%  )
==2581== 
==2581== LL refs:         1,718,101  ( 1,437,069 rd   +    281,032 wr)
==2581== LL misses:         166,689  (    23,034 rd   +    143,655 wr)
==2581== LL miss rate:          0.1% (       0.0%     +        0.6%  )