The API for pwhash_str returns a cstring in the output buffer. These
are null terminated. However, we return the full buffer as a binary
back to Erlang. This means that we have a buffer with 0'es in the end.
The tests take this buffer and passes it back in as is. Hence all the
tests pass. However, it is conceivable that if we write said buffer to
disk somewhere, we are not going to write those 0's out.
When we then load the ASCII-armored Argon2 string into memory again,
it is not 0-terminated as a cstring should be, and this produces
errors all over the place.
The fix is twofold:
* Return the full buffer to Erlang, but use binary:split/2 to create a
subbinary with the relevant part.
* Add a 0 in the end of ASCII Argon2 string before passing it to
libsodium
Since we are looking at pwhashing, and Argon2, we expect the
computational problem to be memory bound. Thus, spending a bit more
work in memory is not going to have any considerable impact on the
speed of this system.
A lot of people who pushed functions they missed have not pushed any
kind of test cases for them. To make sure we have test coverage, I've
marked the functions we have under test and the functions we are still
missing tests for.
Investigation shows erratic behaviour of the CSPRNG. It may block at times
because it needs to go to the kernel and obtain more random information in
the system. Once this happens, you are blocked for quite a long time, but
we don't generally know when this happens. So just execute these directly
on the dirty scheduler.
Rather than creating a set of <<0,…,0>> every time and then throwing it away,
we now reuse a constant. They are small enough they are still small-binaries, but a
process can keep them around and use them whenever it needs to go to the
low-level area.
* Introduce enacl:verify/0 to verify the integrity of the library.
* Always run keypair functions on the dirty schedulers.
* Do not call NIFs when creating zerobytes. Avoid constructing new binary data, but reuse a constant.
* The box/box_open and box_seal/box_seal_open API is so expensive to call there is little overhead in
running it on the DS always.
