Re: [Trisquel-users] Is RdRand in Linux-Libre?

2013-09-13 Thread trinux
[...] RDRAND is still available to user space applications.

[Trisquel-users] Is RdRand in Linux-Libre?

2013-09-12 Thread foxhoundmsx
There's a discussion going around the internet about Intel possibly backdooring Linux with RdRand. Two years ago Linus overrode a decision by the maintainer of /dev/random and made a decision to include a patch by Intel which would make Linux rely blindly on output from RdRand (an

Re: [Trisquel-users] Is RdRand in Linux-Libre?

2013-09-12 Thread foxhoundmsx
Also take a look at this

Re: [Trisquel-users] Is RdRand in Linux-Libre?

2013-09-12 Thread mikko . viinamaki
I believe it is but not as the sole source but added to the mix. This way apparently it doesn't matter if isn't perfectly random.

Re: [Trisquel-users] Is RdRand in Linux-Libre?

2013-09-12 Thread lloyd
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 This is the important point here. If it's the sole source of entropy, that's when it becomes a security risk. As long as it's being mixed with something else, it's pretty safe imo. Maybe there should be (if there isn't already) a piece of

Re: [Trisquel-users] Is RdRand in Linux-Libre?

2013-09-12 Thread mikko . viinamaki
Maybe there should be (if there isn't already) a piece of code that disables RdRand when no other source of entropy is available, to ensure that it's never used as the sole source of entropy. There already is. It's the nordrand parameter to the kernel.

Re: [Trisquel-users] Is RdRand in Linux-Libre?

2013-09-12 Thread gnuser
Hey! Not fair! I spend a week out of here and you guys already go on conspiracy theories without me? =P LOL Ok, seriously now, this is the kind of thing that makes me think free software only, free software only!. Lol. Even so, and as this is an example of, free software does not mean that