tiny.simd.matVecBFloat16Add

1pub fn computeBFloatAdd(2    comptime D: type,3    matrix: []const bfloat.BFloat16,4    vector: []const bfloat.BFloat16,5    add: []const bfloat.BFloat16,6    output: []f32,7) void {8    std.debug.assert(add.len >= output.len);9    computeBFloat(D, matrix, vector, output);10    for (output, add[0..output.len]) |*result, addend| result.* += addend.toF32();11}

1pub const matVecBFloat16Add = matvec.computeBFloatAdd;