Zfh Extension

Versions

1.0.0
State

ratified

Ratification date

== Synopsis

16-bit half-precision binary floating-point instructions compliant with the IEEE 754-2008 arithmetic standard. The Zfh extension depends on the single-precision floating-point extension, F. The NaN-boxing scheme is extended to allow a half-precision value to be NaN-boxed inside a single-precision value (which may be recursively NaN-boxed inside a double- or quad-precision value when the D or Q extension is present).

Instructions

The following instructions are affected by this extension:

fcvt.h.q

Floating-point Convert Quad-precision to Half-precision

fcvt.q.h

Floating-point Convert Half-precision to Quad-precision

fadd.h

No synopsis available

fclass.h

No synopsis available

fcvt.d.h

No synopsis available

fcvt.h.d

Floating-point Convert Double-precision to Half-precision

fcvt.h.l

Floating-point Convert Long to Half-precision

fcvt.h.lu

Floating-point Convert Unsigned Long to Half-precision

fcvt.h.s

Convert half-precision float to a single-precision float

fcvt.h.w

Floating-point Convert Word to Half-precision

fcvt.h.wu

Floating-point Convert Unsigned Word to Half-precision

fcvt.l.h

Floating-point Convert Half-precision to Long

fcvt.lu.h

Floating-point Convert Half-precision to Unsigned Long

fcvt.s.h

Convert single-precision float to a half-precision float

fcvt.w.h

Floating-point Convert Half-precision to Word

fcvt.wu.h

Floating-point Convert Half-precision to Word

fdiv.h

No synopsis available

feq.h

No synopsis available

fle.h

No synopsis available

fleq.h

No synopsis available

flh

Half-precision floating-point load

fli.h

Floating-point Load Immediate Half-precision

flt.h

No synopsis available

fltq.h

No synopsis available

fmadd.h

No synopsis available

fmax.h

No synopsis available

fmaxm.h

No synopsis available

fmin.h

No synopsis available

fminm.h

No synopsis available

fmsub.h

No synopsis available

fmul.h

No synopsis available

fmv.x.h

Move half-precision value from floating-point to integer register

fnmadd.h

No synopsis available

fnmsub.h

No synopsis available

fround.h

No synopsis available

froundnx.h

No synopsis available

fsgnj.h

No synopsis available

fsgnjn.h

No synopsis available

fsgnjx.h

No synopsis available

fsh

Half-precision floating-point store

fsqrt.h

No synopsis available

fsub.h

No synopsis available