[PATCH] Inline C99 math functions

[Wilco Dijkstra]

> Carlos O'Donell wrote:
> On 06/15/2015 01:00 PM, Joseph Myers wrote:
> > On Mon, 15 Jun 2015, Wilco Dijkstra wrote:
> >
> >>> Where are the benchmarks for this?  Please put them in benchtests so
> >>> actual reproducible figures can be given.  That's the standard practice
> >>> for any change being justified on the basis of performance.
> >>
> >> I'll add a benchmark in another patch - it's not trivial as benchtest is not
> >> suitable to accurately time very simple functions, especially when inlined...
> >
> > Well, the benchmark should come first....
> 
> Agreed.
> 
> If it's not trivial to test the performance... then how did you test it?
> 
> How do I test it when reviewing the patch?

Besides checking it was correctly inlining as expected, I ran the usual benchmarks
like SPECFP to check there are no performance regressions due to the math libs.

Adding more inlining is simply a no-brainer, especially when we're talking about 3-4 
instruction functions and not just avoiding a call but also a PLT indirection... 
This provides more than a factor 15 speedup in my microbenchmarks.

Wilco