Any usable SIMD implementation?
9il via Digitalmars-d
digitalmars-d at puremagic.com
Tue Apr 5 04:17:30 PDT 2016
On Tuesday, 5 April 2016 at 10:27:46 UTC, Walter Bright wrote:
> On 4/5/2016 2:03 AM, John Colvin wrote:
> There's a line between trying to standardize everything and
> letting add-on libraries be free to innovate.
>
> Besides, I think it's a poor design to customize the app for
> only one SIMD type. A better idea (I've repeated this ad
> nauseum over the years) is to have n modules, one for each
> supported SIMD type. Compile and link all of them in, then
> detect the SIMD type at runtime and call the corresponding
> module. (This is how the D array ops are currently implemented.)
>
> My experience with command line FPU switches is few users
> understand what they do and even fewer use them correctly.
>
> In fact, I suspect that having a command line FPU switch is too
> global a hammer. A pragma set in just the functions that need
> it might be much better.
>
What wrong for scientist to write `-mcpu=native`?
> -------
>
> In any case, this is not a blocker for getting the library
> designed, built and debugged.
Yes, but this is bad idea to have a set of versions for Phobos,
is not it?
Ilya
More information about the Digitalmars-d
mailing list