Announcement

Collapse
No announcement yet.

Linux 4.4 Crypto Brings Optimized SHA1/SHA256 Intel SHA Extensions

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

  • Linux 4.4 Crypto Brings Optimized SHA1/SHA256 Intel SHA Extensions

    Phoronix: Linux 4.4 Crypto Brings Optimized SHA1/SHA256 Intel SHA Extensions

    Herbert Xu mailed in the crypto subsystem updates this morning for the Linux 4.4 merge window...

    http://www.phoronix.com/scan.php?pag...nux-4.4-Crypto

  • #2
    Right after SHA-1 became obsolete by pretty much everyone and their dog.

    Comment


    • #3
      Originally posted by dimko View Post
      Right after SHA-1 became obsolete by pretty much everyone and their dog.
      I doubt they've spend much effort deliberately optimizing SHA-1 - it's probably a side-effect of the work done for SHA-256. At most, someone probably realised that given half an hour, he could apply the same improvements to the simpler algorithm...

      Comment


      • #4
        Originally posted by Delgarde View Post

        I doubt they've spend much effort deliberately optimizing SHA-1 - it's probably a side-effect of the work done for SHA-256. At most, someone probably realised that given half an hour, he could apply the same improvements to the simpler algorithm...

        Are they really separate algorythms? I remember that back in the day when DES started becoming easy to crack, 3DES became the new standard, which simple ran DES 3x.

        Comment


        • #5
          Yes there are separate algorithms.

          You can find more information about it on Wikipedia, for example compare the operation diagram for 1 round of SHA-2 and SHA-1.
          They are more or less similar, but still separate algorithms.

          SHA-3 is an entirely different class of beasts (a cryptographic sponge function). It's a separate algorithm (Keccak) which got named "SHA-3" because it's offered as a standard and the next in line for using as cryptographic standards after SHA-2.

          3DES was a simple quick and dirty hack to react to the fact that DES was easy to crack. It's really considered as a quick hack (hence all the "double ROT-13 for double security" jokes that arise at the same time). The "real" successor as a cryptographic standard was AES.

          Chaining 2 cryptographic hashing algorithm DOES exist. But it's done for an entirely different reason: it's not for making them more secure, it's used for message authentication (HMAC). The purpose of the 2 rounds of hashing in a HMAC is to make the resulting code imune to Lenght-extension attacks.

          Comment

          Working...
          X