FMMLA (widening, FP16 to FP32)

Half-precision matrix multiply-accumulate to single-precision

This instruction multipliesperforms two fused sums-of-products within each two pairs of adjacent half-precision elements while multiplying the 2×4 matrix of half-precision values held in each 128-bit segment of the first source vector by the 4×2 matrix of half-precision values in the corresponding segment of the second source vector. vector. The intermediate single-precisionsums-of-products results are calculatedrounded asbefore athey sumare of two fusedsummed, sum-of-productsand usingtheir twointermediate pairssum ofis adjacent half-precision values from each of the source vectors, and rounded before accumulation into the 2×2 single-precision matrix of single-precision results in the corresponding segment of the destination vector. This is equivalent to performing a 4-way dot product per destination element.

This instruction is unpredicated.

This instruction is illegal when executed in Streaming SVE mode, unless FEAT_SME_FA64 is implemented and enabled.

SVE2
(FEAT_SVE_F16F32MM)

Decode for this encoding

if !IsFeatureImplemented(FEAT_SVE_F16F32MM) then EndOfDecode(Decode_UNDEF); end; let n : integer = UInt(Zn); let m : integer = UInt(Zm); let da : integer = UInt(Zda);