TW
@enkore Thanks for digging this out. From nowadays perspective, what do you think is good for borg2? - keep blake2b hashing compatible to borg 1.x, so chunk ids stay compatible...
Performance is only inferior if R is slower. But likely it would have to be due to the required properties. E could be done only when needed (at cutting places)....
I didn't suggest using the current 32bit RH as basis, so we would just need a much longer RH. To be used as id hash, it would also need to...
I think I have something now in #8903, please review!
@espadrine What you said was a bit above my head. I'm not a mathematician nor a cryptographer, just a interested nerd (and I did take a crypto course some years...
For borg2, this issue is a `borgstore` issue. The idea is that borg creates the repo starting from nothing and also deletes the repo ending up with nothing. Both of...
@sothix I rebased this on current master now. Likely it will still fail due to the windows-specific `file:` URLs.
Yes, that is not very efficient - but is it really a problem? With compression, that all-zero chunk should be tiny.
See also #1354 and #14.