Author vstinner
Recipients christian.heimes, rhettinger, serhiy.storchaka, vstinner
Date 2013-08-17.23:28:55
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <CAMpsgwbn=8pssCPMNRFdKY3gadWrhUUG4Y6CirW5cN3h2WJwzQ@mail.gmail.com>
In-reply-to <1376780920.98.0.616505973936.issue18771@psf.upfronthosting.co.za>
Content
"+ Cache line benefit:  improves the odds that the adjacent probe will
be on the same cache line (...)
+ Reduced loop overhead:  the second lookup doesn't require a new
computation of the index *i* (we just do a XOR 1) (...)"

With j=11, is the lookup at index 10 in the same cache line? Does a
cache line start at the address of the first lookup, or is it
"rounded" to the size of a cache line?

I read that CPU likes to prefetch data forward, but I didn't know that
they like prefetching backward.
History
Date User Action Args
2013-08-17 23:28:55vstinnersetrecipients: + vstinner, rhettinger, christian.heimes, serhiy.storchaka
2013-08-17 23:28:55vstinnerlinkissue18771 messages
2013-08-17 23:28:55vstinnercreate