Re: block-sha1: improve code on large-register-set machines

From: Brandon Casey <brandon.casey.ctr@nrlssc.navy.mil>
Date: Aug 11, 2009, 21:49 UTC
Message-ID: <s9lHCuz3Fo2GHgUzqSTJyqRY3NBYsra0Ko9LbHd8rN_WaAwuc7IUfw@cipher.nrlssc.navy.mil>
In-Reply-To: <alpine.LFD.2.00.0908111735010.10633@xanadu.home>

Nicolas Pitre wrote:

Show 35 quoted lines

> On Tue, 11 Aug 2009, Brandon Casey wrote:
> 
>> Nicolas Pitre wrote:
>>> On Tue, 11 Aug 2009, Nicolas Pitre wrote:
>>>
>>>> On Tue, 11 Aug 2009, Linus Torvalds wrote:
>>>>
>>>>> On Tue, 11 Aug 2009, Nicolas Pitre wrote:
>>>>>> #define SHA_SRC(t) \
>>>>>>   ({ unsigned char *__d = (unsigned char *)&data[t]; \
>>>>>>      (__d[0] << 24) | (__d[1] << 16) | (__d[2] << 8) | (__d[3] << 0); })
>>>>>>
>>>>>> And this provides the exact same performance as the ntohl() based 
>>>>>> version (4.980s) except that this now cope with unaligned buffers too.
>>>>> The actual object SHA1 calculations are likely not aligned (we do that 
>>>>> object header thing), and if you can't do the htonl() any better way I 
>>>>> guess the byte-based thing is the way to go..
>>> OK, even better: 4.400s.
>>>
>>> This is with this instead of the above:
>>>
>>> #define SHA_SRC(t) \
>>>    ({   unsigned char *__d = (unsigned char *)data; \
>>>         (__d[(t)*4 + 0] << 24) | (__d[(t)*4 + 1] << 16) | \
>>>         (__d[(t)*4 + 2] <<  8) | (__d[(t)*4 + 3] <<  0); })
>>>
>>> The previous version would allocate a new register for __d and then 
>>> index it with an offset of 0, 1, 2 or 3.  This version always uses the 
>>> register containing the data pointer with absolute offsets.  The binary 
>>> is a bit smaller too.
>> In that case, why not change the interface of blk_SHA1Block() so that its
>> second argument is const unsigned char* and get rid of __d and the { } ?
> 
> Because not all architectures care to access the data bytewise.  Please 
> see the original SHA_SRC definition.

You mean this:

   #define SHA_SRC(t) htonl(data[t])

?  Or was there a definition before this one?

I don't follow what you are saying. Are you saying that the following two examples are different?

   unsigned int *data;

   #define SHA_SRC(t) \
      ({   unsigned char *__d = (unsigned char *)data; \
         (__d[(t)*4 + 0] << 24) | (__d[(t)*4 + 1] << 16) | \
         (__d[(t)*4 + 2] <<  8) | (__d[(t)*4 + 3] <<  0); })

and

   unsigned char *data;

   #define SHA_SRC(t) \
      ( (data[(t)*4 + 0] << 24) | (data[(t)*4 + 1] << 16) | \
        (data[(t)*4 + 2] <<  8) | (data[(t)*4 + 3] <<  0) )

-brandon