SVE Instruction List by Dougall Johnson
FCVT: Floating-point convert precision (predicated)
FCVT Zd.D, Pg/M, Zn.H (SVE (SME
svfloat64_t svcvt_f64[_f16]_m(svfloat64_t inactive, svbool_t pg, svfloat16_t op)
128-bit SVE
Set each 64-bit lane of (2) by taking the lowest overlapping 16-bit float from (1) and converting it to a 64-bit float.
256-bit SVE
Set each 64-bit lane of (2) by taking the lowest overlapping 16-bit float from (1) and converting it to a 64-bit float.
512-bit SVE
Set each 64-bit lane of (2) by taking the lowest overlapping 16-bit float from (1) and converting it to a 64-bit float.
Larger sizes
1024-bit SVE
Set each 64-bit lane of (2) by taking the lowest overlapping 16-bit float from (1) and converting it to a 64-bit float.
2048-bit SVE
Set each 64-bit lane of (2) by taking the lowest overlapping 16-bit float from (1) and converting it to a 64-bit float.
Report mistakes or give feedback
Inspired by and based on the x86/x64 SIMD Instruction List by Daytime.