-
Notifications
You must be signed in to change notification settings - Fork 158
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Al/refactor lut #2001
Al/refactor lut #2001
Conversation
703b010
to
057e81b
Compare
c5896c6
to
bc5e468
Compare
backends/tfhe-cuda-backend/cuda/include/integer/integer_utilities.h
Outdated
Show resolved
Hide resolved
bc5e468
to
f27212b
Compare
backends/tfhe-cuda-backend/cuda/include/integer/integer_utilities.h
Outdated
Show resolved
Hide resolved
f27212b
to
ef4e649
Compare
Maybe it would be good to launch the long tests on this PR before merging it, mainly to test the many combinations of the luts |
I'll do a quick run locally with less ops. Since only the C++ side is modified, this does not affect the computation of noise level/degree on the Rust side yet. Next PR should do that. |
ahh ok, then maybe it is not needed, if nothing is using this yet, probably we won't see anything with the tests anyway |
closes: please link all relevant issues
PR content/description
This PR introduces degree / max degree data in the int_radix_lut structure, to prepare for noise tracking in apply_uni/bivariate_lookup_table.
I chose to track max_degree as well because for the many lut case it depends on the number of luts applied.
Check-list: