6 Commits

Author	SHA1	Message	Date
Joe Richey	0445b3e5c1	Use improved algorithm that only uses shifts ... This new algorithm (taken from Wikipedia) only uses shifts, addditions, and subtrations. On my x86_64 machine, the benchmarks are over twice as fast. This also takes num-traits as a dependancy, so that the implementation can be a normal generic function, instread of a macro. Signed-off-by: Joe Richey <joerichey@google.com>	2020-09-08 04:50:50 -07:00
Joe Richey	260148e029	Simplify impl_isqrt macro ... We don't need multiple cases in the macro_rules! Signed-off-by: Joe Richey <joerichey@google.com>	2020-09-07 20:19:44 -07:00
Richard Dodd	9e6ffa6f81	Fix overflow in some edge cases.	2020-01-24 12:25:14 +00:00
Sergey Pepyakin	01bb91452f	Add u128 and i128.	2018-10-30 12:22:24 +01:00
Richard Dodd	5bbfb9187d	Add no_std and fix tests	2018-03-02 21:37:51 +00:00
Richard Dodd	00aabcfe32	First commit	2017-10-23 13:29:42 +01:00