Re: RFC: Run libm tests separately for each function

[Florian Weimer <fw at deneb dot enyo dot de>]

* Joseph Myers:

> On Fri, 17 Feb 2017, Joseph Myers wrote:
>
>> I'd expect this patch to improve test time for sufficiently high 
>> parallelism, where the move from 17 separately built large tests to 1137 
>> smaller tests (on x86_64) allows more parallelism that outweighs the 
>> overheads of running more tests separately from the makefile.  And of 
>> course if you compare with 2.25, that had serialization for the run of 
>> gen-libm-test.pl (which happened unnecessarily as part of the build rather 
>> than only when testing because of the before-compile setting).
>
> A single -j12 run before and after the patch (based on current mainline, 
> so with my move of more functions to auto-libm-test-*) gives timings 
> before
>
> real    0m27.193s
> user    4m27.496s
> sys     0m7.988s
>
> and after
>
> real    0m33.617s
> user    5m11.500s
> sys     0m22.392s
>
> which I consider acceptable (and as noted I'd expect test time to improve 
> for sufficiently high parallelism on systems with enough CPU cores where 
> the previous 17 compilations of 50 MB of code had a serializing effect).

Agreed, these numbers look reasonable.