Announcement

Collapse
No announcement yet.

Due To A GCC Bug, RDRAND Usage Wasn't As Random As Expected

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Due To A GCC Bug, RDRAND Usage Wasn't As Random As Expected

    Phoronix: Due To A GCC Bug, RDRAND Usage Wasn't As Random As Expected

    In a now-fixed bug, the RDRAND/RDSEED instructions for pulling random number generators on modern CPUs may have not been as random as desired when using GCC...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Lol... the funny part was that when Intel* came up with this new (and incredibly useful) hardware assist all the usual suspects moaned about how having a hardware random number generator was some evil Intel-NSA magic backdoor that spies on you 24/7 because numbers.

    Turns out the real security issues had nothing to do with the "evil hardware backdoor" and everything to do with open source software code that had nothing to do with the hardware and that wasn't written correctly.

    * Funny how literally nobody has complained about AMD copying Intel and adding these instructions to RyZen though. Given the usual life-cycle of these conspiracy theories I expect it to go from:

    1. Intel did it, therefore EVIL CONSPIRACY.
    2. AMD copied it: Silence.
    3. A year or two after RyZen has been on the market: Look at this cool random number feature that AMD invented single-handed out of thin air! Thank you for being the only innovative company in the history of mankind AMD!!

    Comment


    • #3
      Originally posted by cuckula View Post
      Turns out the real security issues had nothing to do with the "evil hardware backdoor" and everything to do with open source software code that had nothing to do with the hardware and that wasn't written correctly.
      Nothing to do with the dedicated RNG hardware? It was literally a bug in how said hardware accessed made by the compiler devs. This literally wouldn't have happened if they had stayed with their perfectly functional software RNG.

      1. Intel did it, therefore EVIL CONSPIRACY.
      Intel has a long history of immoral business practices so being on the dole of the government with the biggest surveillance apparatus in the world by a country mile wouldn't exactly be out of character for them.

      2. AMD copied it: Silence.
      First time I've ever heard of it and I'm pretty sure that's not exactly a very unique position on this. It's hard to be up in arms about something you've don't even know about.

      3. A year or two after RyZen has been on the market: Look at this cool random number feature that AMD invented single-handed out of thin air! Thank you for being the only innovative company in the history of mankind AMD!!
      When has AMD ever been given credit for implementing something Intel has already implemented? When they do get credited with something related to x86, like AMD64 or x86_64, it's generally been developed in-house.
      Last edited by L_A_G; 27 July 2017, 09:16 AM.

      Comment


      • #4
        Originally posted by chuckula View Post
        Lol... the funny part was that when Intel* came up with this new (and incredibly useful) hardware assist all the usual suspects moaned about how having a hardware random number generator was some evil Intel-NSA magic backdoor that spies on you 24/7 because numbers.

        Turns out the real security issues had nothing to do with the "evil hardware backdoor" and everything to do with open source software code that had nothing to do with the hardware and that wasn't written correctly.

        * Funny how literally nobody has complained about AMD copying Intel and adding these instructions to RyZen though. Given the usual life-cycle of these conspiracy theories I expect it to go from:

        1. Intel did it, therefore EVIL CONSPIRACY.
        2. AMD copied it: Silence.
        3. A year or two after RyZen has been on the market: Look at this cool random number feature that AMD invented single-handed out of thin air! Thank you for being the only innovative company in the history of mankind AMD!!
        Are you retard? AMD64, integrated GPU, NUMA for X86?

        Comment


        • #5
          Originally posted by chuckula View Post
          * Funny how literally nobody has complained about AMD copying Intel and adding these instructions to RyZen though. Given the usual life-cycle of these conspiracy theories I expect it to go from:

          2. AMD copied it: Silence.
          3. A year or two after RyZen has been on the market: Look at this cool random number feature that AMD invented single-handed out of thin air! Thank you for being the only innovative company in the history of mankind AMD!!
          First of I I believe the instruction set should not be patentable and such for compatible solutions form multiple vendors, Otherwise we ended up with dozens of incompatible CPUs and nothing is interchangeable, like M68k, ARM, MIPS, PowerPC, SPARC, SuperH, etc. pp.

          2nd. Intel copied AMD64, which was way more popular than Intel's failed, expensive, and power hungry IA64 nobody wanted,
          Also IIRC VIA was one of the first to have x86 hardware crypto, I think they called it Padlock or so.

          3. Intel was not that innovative the last years, also initially AMD's hardware virtualization was also way faster, due to Intel not having nested page tables, and also due not virtualizing real mode for BIOS booting and such. On first gen Intel VMX Qemu booting something is really slow until the ReadMode BIOS is finally done. You can literately watch grub paint it's menu pixels, ...

          Comment


          • #6
            Originally posted by chuckula View Post
            Lol... the funny part was that when Intel* came up with this new (and incredibly useful) hardware assist...
            The first HW RNGs definitely date back much longer. There was VIA Padlock long ago and AMD had something on the Geode CPUs (not sure if the NS Geode also had it already). And likely there were other implementations before in ASICs for that purpose.

            Besides, a closed HW thing can never be fully trusted by principle while you can audit a FOSS implementation. That was done, a mistake found and fixed. You'll hardly ever know possible errors (or even backdoors on purpose) in closed HW / FW designs.

            And by the way AMD does push innovations forward, did so in the past and will hopefully do so in the future. Yes, not every idea is 100% original, that is hard to achieve these days. But it is also a matter of bringing an idea to reality and on the market where one can actually buy and use it. AMD, much smaller than intel and suffering from these strange contracts intel had with everyone, was pushing a backward compatible 64bit x86 platform, the "fusion" of CPU and GPU computing, open GPU drivers, just to name some big landmarks.
            Stop TCPA, stupid software patents and corrupt politicians!

            Comment


            • #7
              Checked some info on the matter:
              VIA's first RNGs should have been in the C3/C3-2 (I still have some). Released acc. to wikipedia early 2003.
              Geode LX definitely had a RNG. NS iirc. had the GX and I haven't found RNG infos on that, Geode was sold 2003 to AMD, which continued the GX2 design but quickly developed the LX upgrade.

              And for HW-RNGs in general:
              https://en.wikipedia.org/wiki/Hardwa...tor#Early_work
              (so far you can trust the editors of WP if the did a thorough research, but if you want, every dice is a HW RNG - just not electronic.)
              Stop TCPA, stupid software patents and corrupt politicians!

              Comment


              • #8
                Originally posted by rene View Post
                2nd. Intel copied AMD64, which was way more popular than Intel's failed, expensive, and power hungry IA64 nobody wanted,
                Intel didn't copy x86-64, they licensed it from AMD, just like AMD licenses x86 from Intel. They literally can't exist without eachother at this point.

                As for IA64, it was a FAR superior architecture to x86 and x86-64. The only reason it lost was the emulation penalty to x86 apps. x86-64 was initially higher performing (no 64 bit apps yet) and so won in the market. We'd be in a much better place if we had a CPU arch that was explicitly parallel, rather then stick with x86 based CPUs. [x86 is a HORRID CPU architecture that should have been killed off decades ago]

                Comment


                • #9
                  Originally posted by Adarion View Post
                  Checked some info on the matter:
                  VIA's first RNGs should have been in the C3/C3-2 (I still have some). Released acc. to wikipedia early 2003.
                  Geode LX definitely had a RNG. NS iirc. had the GX and I haven't found RNG infos on that, Geode was sold 2003 to AMD, which continued the GX2 design but quickly developed the LX upgrade.

                  And for HW-RNGs in general:
                  https://en.wikipedia.org/wiki/Hardwa...tor#Early_work
                  (so far you can trust the editors of WP if the did a thorough research, but if you want, every dice is a HW RNG - just not electronic.)
                  Too...many...acronyms...brain...melting.

                  Comment


                  • #10
                    Just your standard first post drive-by anti-AMD trolling, the guy does it at Tech Report too.

                    Comment

                    Working...
                    X