Re: [Qemu-devel] [PATCH 2/2] tcg/optimize: Remember garbage high bits for 32-bit ops

[Richard Henderson <rth@twiddle.net>]

On 05/23/2014 12:46 PM, Paolo Bonzini wrote:
>> @@ -166,11 +166,18 @@ static void tcg_opt_gen_mov(TCGContext *s, int
>> op_index, TCGArg *gen_args,
>>                              TCGOpcode old_op, TCGArg dst, TCGArg src)
>>  {
>>      TCGOpcode new_op = op_to_mov(old_op);
>> +    tcg_target_ulong mask;
>>
>>      s->gen_opc_buf[op_index] = new_op;
>>
>>      reset_temp(dst);
>> -    temps[dst].mask = temps[src].mask;
>> +    mask = temps[src].mask;
>> +    if (TCG_TARGET_REG_BITS > 32 && new_op == INDEX_op_mov_i32) {
>> +        /* High bits of the destination are now garbage.  */
> 
> Or they are zero on x86_64... perhaps this could be some kind of TCG target hook.

Only if we actually issued a "mov" insn.

If the register allocator decided to adjust its tables instead, we won't issue
an insn, at which point we really do have garbage in the high bits.  I don't
think we should be thinking about the target at all at this point.


r~