See "FMLALT (vectors, FP8 to FP16)" in the exploration tools

FMLALT (vectors, FP8 to FP16): 8-bit floating-point multiply-add to half-precision (top)

FMLALT Zda.H, Zn.B, Zm.B (SVE2+FP8FMA+NS (SSVE-FP8FMA

128-bit SVE

For each odd 8-bit float calculate (1) * (2), and add that to the 16-bit float from (3), then set (4) to the result. The FP8 format for each 8-bit source operand is selected independently by FPMR.

256-bit SVE

For each odd 8-bit float calculate (1) * (2), and add that to the 16-bit float from (3), then set (4) to the result. The FP8 format for each 8-bit source operand is selected independently by FPMR.

512-bit SVE

For each odd 8-bit float calculate (1) * (2), and add that to the 16-bit float from (3), then set (4) to the result. The FP8 format for each 8-bit source operand is selected independently by FPMR.

Larger sizes

1024-bit SVE

For each odd 8-bit float calculate (1) * (2), and add that to the 16-bit float from (3), then set (4) to the result. The FP8 format for each 8-bit source operand is selected independently by FPMR.

2048-bit SVE

For each odd 8-bit float calculate (1) * (2), and add that to the 16-bit float from (3), then set (4) to the result. The FP8 format for each 8-bit source operand is selected independently by FPMR.

Report mistakes or give feedback
Inspired by and based on the x86/x64 SIMD Instruction List by Daytime.