You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I noticed two potential limitations(maybe intended?) in the current Poseidon hash function implementation:
The size of the rate part is currently fixed at (the size of the sponge state - 1) as shown in here:
While this configuration may provide sufficient security for large prime fields (~255 bits), it might not be adequate for smaller prime fields (e.g., Goldilocks). I think we should allow developers to configure the size of the capacity part to ensure appropriate security levels for various applications and prime field sizes.
Also the current implementation only returns one element from the rate part during the squeezing phase. It would be beneficial to enable squeezing multiple elements from the rate part, with additional permutations if necessary, to accommodate different use cases.
I recall our previous discussion on this topic; however, do you believe it is worthwhile to pursue further? @kilic
The text was updated successfully, but these errors were encountered:
I noticed two potential limitations(maybe intended?) in the current Poseidon hash function implementation:
The size of the rate part is currently fixed at (the size of the sponge state - 1) as shown in here:
While this configuration may provide sufficient security for large prime fields (~255 bits), it might not be adequate for smaller prime fields (e.g., Goldilocks). I think we should allow developers to configure the size of the capacity part to ensure appropriate security levels for various applications and prime field sizes.
Also the current implementation only returns one element from the rate part during the squeezing phase. It would be beneficial to enable squeezing multiple elements from the rate part, with additional permutations if necessary, to accommodate different use cases.
I recall our previous discussion on this topic; however, do you believe it is worthwhile to pursue further? @kilic
The text was updated successfully, but these errors were encountered: