more initial non-randomness
Closed this issue · 0 comments
cladisch commented
The problem of issue #1 is still there (in non-encryption mode, the first actually output block consists of zeros only).
Commit cc8d0ff sets "output_ready" for every block, even for one that has failed the FIPS tests. So of the first block fails these tests (which is quite common), the first output block still is empty.
The proposed patch in issue #1 would have fixed this completely.