On Dec 12, 2011, at 10:29 PM, Marsh Ray wrote:

> On 12/12/2011 12:55 PM, Yoav Nir wrote:
>> 
>> On Dec 12, 2011, at 5:52 PM, Marsh Ray wrote:
>>> 
>>> It's already somewhat ambiguous now that NIST has
>>> defined SHA[-2]-512/256.
>>> 
>>> http://csrc.nist.gov/publications/PubsDrafts.html#fips-180-4
>> 
>> Then that is what it must be called: "sha2-512/256". I think that's a legal 
>> string in HTTP headers.
>> 
>> Supposedly this is faster on 64-bit applications. I wonder if that is true 
>> in practice.
> 
> SHA-2-512/256 should perform identically to plain SHA-2-512. It's the 
> same function only with a different IV.

Yes, and the claim in the original paper was that SHA2-512 was faster on 64-bit 
systems than SHA2-256. I haven't verified this.

> This site is dedicated to benchmarking hash functions:
> http://bench.cr.yp.to/results-hash.html
> 
> It doesn't appear to be faster than, say, SHA-1, SHA-2-512 but does not 
> appear to be subject to the same attacks either.
> 
>> So far, I have seen no implementations of this hash function.
> 
> I made one for use at work, but it's only a minor adjustment to the RFC 
> 6234 SHA-2-512 reference code.

I did that as well.

_______________________________________________
websec mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/websec

Reply via email to