Can you name applications for CLMUL that are not CRC, GCM, hashing/rng, or Erasure Code? I'm trying to create a list of possible applications. #RISCV #Bitmanip #Followerpower cc @rygorous @geofflangdale @alt_kia @lemire @rrika9
-
Show this thread
-
CLMUL a number with itself inserts zeroes between each input bit. Useful if you want to construct morton-codes. They are useful to index into large 2D arrays because they are very cache friendly.
1 reply 0 retweets 8 likes -
Thanks! I've now added that as clmul(h) use-case (and "zip" use-case, which is the more obvious choice for morton-codes within xbitmanip) to the current draft document at https://raw.githubusercontent.com/cliffordwolf/xbitmanip/master/xbitmanip-draft.pdf …, and added you as contributor, and added the following invariant to my tests.pic.twitter.com/32dWnYWLTh
1 reply 0 retweets 1 like -
I have another one. Afaik you can calculate the Number Theoretic Transform in GF2 using clmul. This gives you most of the goodies a FFT will, but without ever running into roundig Errors. You can use this for example to do fast bignum multiplications.
1 reply 0 retweets 2 likes -
Do you have a reference for this? (Googled a bit but didn't find something that looks like that, maybe I just used the wrong search terms? I guess the fast bignum part refers to the Schoenhage-Strassen algorithm. But I can't seem to find anything specifically re NTT using CLMUL.)
1 reply 0 retweets 0 likes -
Here is a paper using NTT over a Galois Field doing image filtering: https://www.computer.org/csdl/trans/tc/1977/09/01674935.pdf … They don't use GF(2) which clmul does natively, but GF(45 X 229 + 1) instead. I'm pretty sure that clmul can be used to compute these, but it's not exactly my area of expertise :-)
1 reply 0 retweets 2 likes -
Ah, no. dug a bit deeper. You can't use clmul for that. It's plain modulo-prime arithmetic that they are using. :-/
1 reply 0 retweets 1 like -
You absolutely can use clmul for Galois Field arithmetic, of course, that's why it was introduced. You can do the modulo reduction in two clmul, e.g., see https://arxiv.org/abs/1503.03465
1 reply 0 retweets 1 like -
Forgive my ignorance and maybe it's in your paper and I just don't see it, but it's not obvious at all to me how to use that to implement GF(p) arithmetic (not GF(2^n) but a non-2 prime field). Nils: 45*299 + 1 = 13456 = 2^4 * 29^2 afaict. So no such GF exists?
1 reply 0 retweets 1 like
Nils: Ah! It's 45*2^29+1. 
Yeah, that's a prime field, everything alright there then.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.