-
Notifications
You must be signed in to change notification settings - Fork 143
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
feat(integer): improve scalar_mul #517
Conversation
@slab-ci cpu_fast_test |
8954b7b
to
3c8088d
Compare
@slab-ci cpu_fast_test |
3c8088d
to
77b3662
Compare
@slab-ci cpu_fast_test |
77b3662
to
9225d65
Compare
@slab-ci cpu_fast_test |
9225d65
to
ab3037c
Compare
@slab-ci cpu_fast_test |
ab3037c
to
f44f0ad
Compare
@slab-ci cpu_fast_test |
@slab-ci cpu_fast_test |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I realize my comment is on some "old" code, but I could not help but notice it may not be quite right ?
Pull Request has been approved 🎉 |
This changes the algorithm for scalar_mul. The new algorithm allows to remove a lot of work. For small precisions (16, 32, 64) the gains are in range 5%-10% for higher precisions the gains are 25%-50%. This also changes the mul to use the functions that sums many clean ciphertexts in parallel. For mul, there is only a 5%-10% improvements for 128bits and 256bits mul.
f44f0ad
to
006cc63
Compare
@slab-ci cpu_fast_test |
what's the diff on your last push ? |
Just rebased on top of the fix to the signed_div_rem test in the case of doing 0/0 |
Pull Request has been approved 🎉 |
PR content/description
This changes the algorithm for scalar_mul.
The new algorithm allows to remove a lot of work.
For small precisions (16, 32, 64) the gains are in range 5%-10% for higher precisions the gains are 25%-50%.
This also changes the mul to use the functions that sums many clean ciphertexts in parallel. For mul, there is only a 5%-10% improvements for 128bits and 256bits mul.