News for hackers. Stuff that matters.
Re:Galois field instructions on 2021 CPUs by corsix on Sunday November 06, 2022 @11:33AM (#33494316)
The next article in blog order is one application: https://www.corsix.org/content/reed-solomon-for-software-rai...

Another application is crypto: the SubBytes step of AES maps very neatly onto gf2p8affineinvqb, so algorithms that are similar to AES but not exactly AES could make use of gf2p8affineinvqb

Re:Galois field instructions on 2021 CPUs by salicideblock on Sunday November 06, 2022 @01:05PM (#33495357)
Expanding on this,a very nice property of Galois Counter Mode (GCM) for AES is that encrypting one block does not require the previous block to be encrypted, like in AES-CBC.

This means that AES-GCM can take advantage of data parallelism and there are big speedups in threaded and pipelined CPUs.

In short, you can get big latency and throughout gains by using AES-GCM over AES-CBC.

Re:Galois field instructions on 2021 CPUs by nwellnhof on Sunday November 06, 2022 @01:57PM (#33495847)
> algorithms that are similar to AES but not exactly AES

The SM4 cipher, for example: https://en.wikipedia.org/wiki/SM4_%28cipher%29