News for hackers. Stuff that matters.
Re:The many ways of converting FP32 to FP16 by Joker_vD on Sunday April 23, 2023 @11:36AM (#35677144)
> Sadly, at some point, new hardware makes software tricks obsolete. On x86, said new hardware was the F16C instruction set, which (amongst other things) adds a VCVTPS2PH instruction for converting (a vector of) FP32 to FP16.

"Sadly"? That's subjective, I guess: I personally feel that the less "softfloat" code there is in use around, the better. All this twisted bitshifting/juggling with conditional propagation is pretty well suited for baking into sillicon.

Re:The many ways of converting FP32 to FP16 by kzrdude on Sunday April 23, 2023 @01:05PM (#35678106)
The converse is also deserving of a 'sadly': the new hardware features can be widely available but still hard to use since they are not going to be used by platform defaults when you compile your program.
Re:The many ways of converting FP32 to FP16 by stephencanon on Sunday April 23, 2023 @03:37PM (#35679732)
F16C has been available for well over a decade (it was introduced in the Ivybridge uArch), and the equivalent conversions are supported on all ARMv7 CPUs since around the same time (and all ARMv8 CPUs unconditionally); most software should be able to safely assume their availability by now.

There exist some niches where this doesn’t hold, but that’s why we have software fallbacks.

Re:The many ways of converting FP32 to FP16 by corsix on Sunday April 23, 2023 @04:40PM (#35680366)
The first niche that came to mind was x86 code running under Rosetta 2; despite ARM having an equivalent to F16C, Rosetta 2 doesn’t translate AVX, and F16C doesn’t have a non-AVX encoding.
Re:The many ways of converting FP32 to FP16 by stephencanon on Sunday April 23, 2023 @05:52PM (#35681034)
Indeed. Worth noting that Accelerate.framework provides fast and correct bulk f16 <-> f32 conversions as `vImageConvert_Planar16FtoPlanarF` and `vImageConvert_PlanarFtoPlanar16F`, and that the arm conversion instructions are unconditionally available for apps that compile for arm64 (they're part of the base ARMv8 ISA), so any _new_ code shouldn't need to worry about this.
Re:The many ways of converting FP32 to FP16 by diydsp on Sunday April 23, 2023 @01:19PM (#35678263)
Im solving this problem by writing fp16 routines for 6502 :)
Re:The many ways of converting FP32 to FP16 by fuzzfactor on Sunday April 23, 2023 @11:27PM (#35683349)
In my case it wasn't exactly conversion but I had pages of FORTRAN algorithm based on precision numerical constants in FP32, and the outcome needed to be mimicked in 8-bit BASIC using 0.5K of memory for program plus data.

Not enough memory to even store the text of the FORTRAN code.

Integer math was required :/