Re: PATCH: Round 2 of octeon rng

2013-10-23 Thread Ted Unangst
On Tue, Oct 22, 2013 at 22:05, William Orr wrote: I guess I misunderstood, as I thought that /dev/random dumped the entropy pool, and that /dev/arandom put the random data through a stream cipher so that grabbing random data would never block. That was true some time ago, but since at least

Re: PATCH: Round 2 of octeon rng

2013-10-23 Thread Paul Irofti
Hi William, I have an almost identical diff in my tree for a driver for octeon's RNG. On which machines did you test this? I wrote mine for DSR-500. But I remember the reads had some hick-ups. Since why I did not commit the diff until now.

Re: PATCH: Round 2 of octeon rng

2013-10-23 Thread William Orr
On Oct 23, 2013, at 4:38 AM, Paul Irofti p...@irofti.net wrote: Hi William, I have an almost identical diff in my tree for a driver for octeon's RNG. On which machines did you test this? I wrote mine for DSR-500. But I remember the reads had some hick-ups. Since why I did not commit

Re: PATCH: Round 2 of octeon rng

2013-10-23 Thread William Orr
On Oct 22, 2013, at 11:54 PM, Ted Unangst t...@tedunangst.com wrote: On Tue, Oct 22, 2013 at 22:05, William Orr wrote: I guess I misunderstood, as I thought that /dev/random dumped the entropy pool, and that /dev/arandom put the random data through a stream cipher so that grabbing random

PATCH: Round 2 of octeon rng

2013-10-22 Thread William Orr
Hi again tech@ This is my second attempt at a patch to add support for the octeon's onboard rng. I've fixed all of the concerns (ISC license, wrong #define, comment removal) and I've also come bearing statistics on the quality of the entropy. I dd'd 512M of /dev/random and ran the ent from

Re: PATCH: Round 2 of octeon rng

2013-10-22 Thread Ted Unangst
On Tue, Oct 22, 2013 at 18:31, William Orr wrote: You'll notice that there's no significant difference between the output of the two rngs. However, with octrng the dd completed in under a minute (more entropy in pool). Without, it took several minutes. If you want time output, I can add that

Re: PATCH: Round 2 of octeon rng

2013-10-22 Thread William Orr
On Oct 22, 2013, at 9:06 PM, Ted Unangst t...@tedunangst.com wrote: On Tue, Oct 22, 2013 at 18:31, William Orr wrote: You'll notice that there's no significant difference between the output of the two rngs. However, with octrng the dd completed in under a minute (more entropy in pool).

Re: PATCH: Round 2 of octeon rng

2013-10-22 Thread Theo de Raadt
I can do this again with time, but pulling data from /dev/random took significantly longer without my patch than with it. That is not possible.