==2599== Cachegrind, a cache and branch-prediction profiler
==2599== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2599== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2599== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/ltmain.sh
==2599== Parent PID: 2592
==2599== 
--2599-- warning: L3 cache found, using its data for the LL simulation.
==2599== brk segment overflow in thread #1: can't grow to 0x482b000
==2599== (see section Limitations in user manual)
==2599== NOTE: further instances of this message will not be shown
==2599== 
==2599== I   refs:      112,718,331
==2599== I1  misses:      1,004,095
==2599== LLi misses:          8,761
==2599== I1  miss rate:        0.89%
==2599== LLi miss rate:        0.01%
==2599== 
==2599== D   refs:       53,184,105  (27,628,155 rd   + 25,555,950 wr)
==2599== D1  misses:        670,944  (   389,784 rd   +    281,160 wr)
==2599== LLd misses:        157,143  (    13,975 rd   +    143,168 wr)
==2599== D1  miss rate:         1.3% (       1.4%     +        1.1%  )
==2599== LLd miss rate:         0.3% (       0.1%     +        0.6%  )
==2599== 
==2599== LL refs:         1,675,039  ( 1,393,879 rd   +    281,160 wr)
==2599== LL misses:         165,904  (    22,736 rd   +    143,168 wr)
==2599== LL miss rate:          0.1% (       0.0%     +        0.6%  )