[Octave-bug-tracker] [bug #58800] BIST for rng sometimes fails

[Markus Mützel]

Follow-up Comment #11, bug #58800 (project octave):

Reading on a little bit more about it, I believe it is save to replace reading
from /dev/urandom with std::random_device on Linux. It should do essentially
the same.
Since the standard doesn't guarantee that the random numbers generated by
random_device are non-deterministic (or cryptographically secure), I moved the
other sources to the front.
AFAICT, random_device is a "good" source of entropy on Linux and Windows.
Worst case on other platforms might be that it doesn't add any more entropy.
But like Michael Leitner pointed out, the clock should be enough anyway.

The patch also adds the PID to the initialization vector like Michael Leitner
suggested.

It looks like the changes didn't break the rng on Linux or Windows for me. It
would be interesting to see if it fixes the original issue.

(file #49536)
    _______________________________________________________

Additional Item Attachment:

File name: bug58800_random_device_v2.patch Size:2 KB
   
<https://file.savannah.gnu.org/file/bug58800_random_device_v2.patch?file_id=49536>



    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?58800>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/