Bicliques preimage attack, is it a worry?

[David Rabahy]

http://eprint.iacr.org/2011/286.pdf

[gmaxwell]

No, more certification weaknesses (around 2^256 work) on reduced round versions.

[David Rabahy]

Ah, SHA-256 is 64 rounds, whereas the paper in question talks about of an attack at 45 rounds.  I gather it is not just a matter of working harder the same way to get to 46 or more rounds but rather novel enhancements are required if it is even possible.

[David Rabahy]

One wonders if the Bitcoin reference implementation is built upon one of the SHS validated http://csrc.nist.gov/groups/STM/cavp/documents/shs/shaval.htm SHA-256 implementations.

[David Rabahy]

The referenced paper is obviously a public attack, so to speak.  One wonders what the state of the art is non-publicly.   I suppose the paper might be as good as it gets at this point.