Quoting Marko Vojinovic vvmarko@gmail.com:
On Thursday, October 28, 2010 21:36:58 you wrote:
On Thursday, October 28, 2010 19:27:16 William Case wrote:
How does the cpu search and find stuff?
There is a huge amount of searching and finding of text in memory, conditional statements requiring comparisons, and the use of entry points but not exact addresses from within both kernel space and user space. It has occurred to me that a there is necessarily a lot of physical or bit comparing going on. Too much, I would think, to keep dumping a search criteria into a cpu register and then replacing the contents of a second register from a block of memory until one matches.
Believe it or not, in a nutshell that's exactly what is happening.
By the way, the sheer inefficiency of that searching algorithm (ie. what you are complaining about) is _precisely_ one of the reasons why quantum computers are so interesting (the other reason is the number factorization into primes). But that's going a bit off-topic, I guess... ;-)
Best, :-) Marko
surely the kernel folks know about boyer-moore and other much better algorithms
http://en.wikipedia.org/wiki/Boyer%E2%80%93Moore_string_search_algorithm
Dave
-- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines