create DirtyOutOfPlaceMontgomeryModMul #1395
Conversation
| target += Tm | ||
| target >>= self.window_size | ||
| qrom_indices = (qrom_indices << self.window_size) | m | ||
| return target, qrom_indices |
There was a problem hiding this comment.
this is now duplicate code, right? See my other comment about how we can make the classical action more readable / useful
There was a problem hiding this comment.
no .. this is part of the classical action function for DirtyOutOfPlaceMontgomeryModMul
There was a problem hiding this comment.
this code is nearly identical to SingleWindowModMul.on_classical_vals. Can you explain the distinction (if any) between the two, preferably in code comments
There was a problem hiding this comment.
the main difference is that SingleWindowModMul.on_classical_vals operates on bitarrays. this methods operates on ints.
added a comment.
NoureldinYosri
force-pushed
the
mod_mul
branch
from
9d731ff to
9a99dcc
Compare
mpharrigan
left a comment
mpharrigan
left a comment
There was a problem hiding this comment.
can you please address my remaining new comments, then lgtm
3 participants
Originall created in #1371. I had to recreate the PR because of issues with my clone of Qualtran.
I addressed the comments on the original PR and switched from QROM to QROAMClean which leads to a better Toffoli complexity than in table 8 in Litinski https://arxiv.org/abs/2306.08585.
This implementation follows appendix C4 in Gouzien https://arxiv.org/abs/2302.06639 and has Toffoli complexity of$2.25n^2+8.25n-1$ which is better than the $2.25n^2+9n$ in table 8 in Litinski https://arxiv.org/abs/2306.08585