Re: [PATCH 1/3] crypto: X25519 low-level primitives for ppc64le.

[Danny Tsen <dtsen@linux.ibm.com>]

See inline.

On 5/15/24 4:06 AM, Andy Polyakov wrote:
> Hi,
>
>> +SYM_FUNC_START(x25519_fe51_sqr_times)
>> ...
>> +
>> +.Lsqr_times_loop:
>> ...
>> +
>> +    std    9,16(3)
>> +    std    10,24(3)
>> +    std    11,32(3)
>> +    std    7,0(3)
>> +    std    8,8(3)
>> +    bdnz    .Lsqr_times_loop
>
> I see no reason for why the stores can't be moved outside the loop in 
> question.
>
Yeah.  I'll fix it.


>> +SYM_FUNC_START(x25519_fe51_frombytes)
>> +.align    5
>> +
>> +    li    12, -1
>> +    srdi    12, 12, 13    # 0x7ffffffffffff
>> +
>> +    ld    5, 0(4)
>> +    ld    6, 8(4)
>> +    ld    7, 16(4)
>> +    ld    8, 24(4)
>
> Is there actual guarantee that the byte input is 64-bit aligned? While 
> it is true that processor is obliged to handle misaligned loads and 
> stores by the ISA specification, them being inefficient doesn't go 
> against it. Most notably inefficiency is likely to be noted at the 
> page boundaries. What I'm trying to say is that it would be more 
> appropriate to avoid the unaligned loads (and stores).

Good point.  Maybe I can handle it with 64-bit aligned for the input.

Thanks.


>
> Cheers.
>