On 4/30/20 11:09 AM, Peter Maydell wrote:
> These don't use do_3same() because they want to
> operate on VFP double registers, whose offsets are different from the
> neon_reg_offset() calculations do_3same does.
Actually, no, it's an around the bush way of computing the same register offset.
vfp_reg_offset(true, reg)
-> vfp.zregs[reg >> 1].d[reg & 1];
neon_reg_offset(reg, 0)
-> vfp_reg_offset(false, 2 * reg + 0)
-> vfp.zregs[(2 * reg) >> 2].d[((2 * reg) >> 1) & 1]
+ ((2 * reg) & 1) * offsetof(lower/upper)
-> vfp.zregs[reg >> 1].d[reg & 1] + 0
r~
