Answer by geza for How can I accurately benchmark unaligned access speed on...
I'm putting my little bit improved benchmark here. Still measures throughput only (and only unaligned offset 1). Based on the other answers, I've added measuring 64- and 4096-byte splits.For 4k splits,...
View ArticleAnswer by harold for How can I accurately benchmark unaligned access speed on...
Testing 64-bit loads for various offsets (code below), my raw results on Haswell are:aligned L: 4.01115 T: 0.500003ofs1 L: 4.00919 T: 0.500003ofs2 L: 4.01494 T: 0.500003ofs3 L: 4.01403 T: 0.500003ofs7...
View ArticleAnswer by Peter Cordes for How can I accurately benchmark unaligned access...
Timing method. I probably would have set it up so the test was selected by a command-line argument, so I could time it with perf stat ./unaligned-test, and get perf counter results instead of just...
View ArticleHow can I accurately benchmark unaligned access speed on x86_64?
In an answer, I've stated that unaligned access has almost the same speed as aligned access a long time (on x86/x86_64). I didn't have any numbers to back up this statement, so I've created a benchmark...
View Article
