std.benchmarking and getting rid of TickDuration

2015-10-05 Thread Jonathan M Davis via Digitalmars-d
I just created a PR aimed at replacing the few pieces of Phobos that use TickDuration in their API, and it involves creating the std.benchmark module to hold the updated benchmarking functions, since they can't be overloads of the existing ones in std.datetime given that return types are

Re: std.benchmarking and getting rid of TickDuration

2015-10-05 Thread Jonathan M Davis via Digitalmars-d
LOL. I got the title wrong - std.benchmarking instead of std.benchmark. Oh well... - Jonathan M Davis

Re: std.benchmarking and getting rid of TickDuration

2015-10-05 Thread Marco Leise via Digitalmars-d
Should the examples have `pragma(inline, false)` on the benchmarked functions? I'm not so worried about inlining as I am about const folding the benchmarked expressions away. Concerning the module review I tend to agree with you. It would only be ... well if I said in std.datetime the functions

Re: std.benchmarking and getting rid of TickDuration

2015-10-05 Thread Jonathan M Davis via Digitalmars-d
On Monday, 5 October 2015 at 15:51:58 UTC, Marco Leise wrote: Should the examples have `pragma(inline, false)` on the benchmarked functions? I'm not so worried about inlining as I am about const folding the benchmarked expressions away. I'm not sure that it's a good idea to start pasting