On Sat, Feb 4, 2012 at 4:12 PM, Bert Huijben <b...@qqmail.nl> wrote: > We already assume Sha-1 is unique in wc-Ng, so introducing sha256 would > not help.
Well, it wouldn't be difficult to bump the wc version and recompute sha256s for all the pristines. I'm not advocating that course of action, just saying it would be possible. This does bring up a good point, though. What kind of forward compatibility, if any, do we need in Ev2 to support a checksum migration? > It is also much more expensive to obtain then sha1 and md5. Actually, sha256 is only about 35% slower than sha1, whereas sha1 is more than 65% slower than md5 (Source: http://www.cryptopp.com/benchmarks.html). I don't know if apr has a sha256 implementation, but it wouldn't be hard to find one. -Hyrum -- uberSVN: Apache Subversion Made Easy http://www.uberSVN.com/
