github-actions[bot]
commented
2 months ago Valgrind Results:
==3922== Cachegrind, a cache and branch-prediction profiler
==3922== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==3922== Using Valgrind-3.18.1 and LibVEX; rerun with -h for copyright info
==3922== Command: target/release/snailtracer
==3922==
--3922-- warning: L3 cache found, using its data for the LL simulation.
Running snailtracer example!
elapsed: 1.839970174s
==3922==
==3922== I refs: 427,733,742
==3922== I1 misses: 3,275
==3922== LLi misses: 2,827
==3922== I1 miss rate: 0.00%
==3922== LLi miss rate: 0.00%
==3922==
==3922== D refs: 204,081,723 (132,272,719 rd + 71,809,004 wr)
==3922== D1 misses: 339,673 ( 199,735 rd + 139,938 wr)
==3922== LLd misses: 137,644 ( 4,352 rd + 133,292 wr)
==3922== D1 miss rate: 0.2% ( 0.2% + 0.2% )
==3922== LLd miss rate: 0.1% ( 0.0% + 0.2% )
==3922==
==3922== LL refs: 342,948 ( 203,010 rd + 139,938 wr)
==3922== LL misses: 140,471 ( 7,179 rd + 133,292 wr)
==3922== LL miss rate: 0.0% ( 0.0% + 0.2% )
if the artificially big number is set for the block number it will have different values on 32 and 64 systems if the number is 2^32 < N < 2^64