Announcement

Collapse
No announcement yet.

Linux 6.1 Continues Improving The RNG & Crypto Code

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

  • Linux 6.1 Continues Improving The RNG & Crypto Code

    Phoronix: Linux 6.1 Continues Improving The RNG & Crypto Code

    The random number generator "RNG" and crypto subsystem pull requests have already been submitted for the Linux 6.1 merge window...

    https://www.phoronix.com/news/Linux-6.1-RNG-Crypto

  • #2
    It's nice to see all these improvements coming through, but even when numbers are provided such as a 50% improvement in the speed of calls to random, it's hard to grasp what that actually means for a user.

    Could we possibly see some benchmarks comparing throughput of wireguard, or another task that might be heavily influenced by this comparing now/6.1 and a year or two ago?

    Comment


    • #3
      Someone please peer-review all commits affecting the RNG.

      Comment


      • #4
        Originally posted by uid313 View Post
        Someone please peer-review all commits affecting the RNG.
        I'd certainly welcome more peer review. But also, most of the larger RNG changes have somebody's `Reviewed-by:` in them. I'm not sure whether you were aware of this when making your comment. Perhaps instead what you meant to say was, "I don't trust any changes to the RNG, ever, and so I prefer to keep using the old broken code," in which case, you're certainly entitled to that preference, but I'm not sure it's a very popular one.

        Comment


        • #5
          Originally posted by zx2c4 View Post

          I'd certainly welcome more peer review. But also, most of the larger RNG changes have somebody's `Reviewed-by:` in them. I'm not sure whether you were aware of this when making your comment. Perhaps instead what you meant to say was, "I don't trust any changes to the RNG, ever, and so I prefer to keep using the old broken code," in which case, you're certainly entitled to that preference, but I'm not sure it's a very popular one.
          Well said, Mr. Donenfeld. I, for one, appreciate your work to improve the kernel's RNG. Thank you for not being afraid to change things significantly in order to improve both the quality and speed of the subsystem.

          Comment


          • #6
            Originally posted by zx2c4 View Post

            I'd certainly welcome more peer review. But also, most of the larger RNG changes have somebody's `Reviewed-by:` in them. I'm not sure whether you were aware of this when making your comment. Perhaps instead what you meant to say was, "I don't trust any changes to the RNG, ever, and so I prefer to keep using the old broken code," in which case, you're certainly entitled to that preference, but I'm not sure it's a very popular one.
            I wasn't aware of this.

            Thank you for your contributions and daring to touch the RNG.

            Comment


            • #7
              Originally posted by zx2c4 View Post

              I'd certainly welcome more peer review. But also, most of the larger RNG changes have somebody's `Reviewed-by:` in them. I'm not sure whether you were aware of this when making your comment. Perhaps instead what you meant to say was, "I don't trust any changes to the RNG, ever, and so I prefer to keep using the old broken code," in which case, you're certainly entitled to that preference, but I'm not sure it's a very popular one.
              i mean, its a comment section on a forum. he didnt say "oh no you are changing this code, that makes me angry and i wish you werent doing it." to the contrary he was just randomly noting that rng is the sort of thing that could always potentially be targeted by bad actors, which is both true and a totally normal thing to chat about in a forum thread. beyond that, updates to this code are awesome and appreciated, which he also just said, even after you reacted like a triggered little girl for no reason. not cool.

              and since you are so fond of telling people what they must have meant to say, heres what YOU must have meant to say:

              "hi, thanks for your interest! you are right to have that concern, heres what i can tell you about it: ..."

              or just dont say anything at all.
              Last edited by quaz0r; 05 October 2022, 11:51 PM.

              Comment

              Working...
              X