Skip to content
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

Faster AVX512 #405

Merged
merged 1 commit into from
Feb 26, 2024
Merged

Faster AVX512 #405

merged 1 commit into from
Feb 26, 2024

Conversation

spapinistarkware
Copy link
Contributor

@spapinistarkware spapinistarkware commented Feb 25, 2024

This change is Reviewable

Copy link
Contributor Author

Current dependencies on/for this PR:

This stack of pull requests is managed by Graphite.

@spapinistarkware spapinistarkware changed the title Faster AVX256 Faster AVX512 Feb 25, 2024
@spapinistarkware spapinistarkware force-pushed the spapini/02-25-Faster_AVX256 branch 2 times, most recently from 6659b9f to cf22cc2 Compare February 25, 2024 19:01
@ilyalesokhin-starkware
Copy link
Collaborator

src/core/fields/avx512_m31.rs line 101 at r2 (raw file):

            let rm_m_p = _mm512_sub_epi32(rm, P);

            Self(_mm512_min_epu32(rm, rm_m_p))

this is add_mod_pm right?

Code quote:

            let rm = _mm512_add_epi32(rm_l, rm_h);
            let rm_m_p = _mm512_sub_epi32(rm, P);

            Self(_mm512_min_epu32(rm, rm_m_p))

Copy link
Collaborator

@ilyalesokhin-starkware ilyalesokhin-starkware left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 4 files at r2.
Reviewable status: 1 of 4 files reviewed, 1 unresolved discussion (waiting on @shaharsamocha7 and @spapinistarkware)

Copy link
Contributor Author

@spapinistarkware spapinistarkware left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: 1 of 4 files reviewed, 1 unresolved discussion (waiting on @ilyalesokhin-starkware and @shaharsamocha7)


src/core/fields/avx512_m31.rs line 101 at r2 (raw file):

Previously, ilyalesokhin-starkware wrote…

this is add_mod_pm right?

Yes, but it's a different PR stack. The intention is to use this one in the butterfly code. I better move the docs here though...

@spapinistarkware spapinistarkware force-pushed the spapini/02-25-Faster_AVX256 branch from cf22cc2 to 591f7ca Compare February 26, 2024 13:00
Copy link
Contributor Author

@spapinistarkware spapinistarkware left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: 0 of 4 files reviewed, 3 unresolved discussions (waiting on @ilyalesokhin-starkware and @shaharsamocha7)


src/core/fields/avx512_m31.rs line 32 at r2 (raw file):

    }

    pub fn reduce(self) -> M31AVX512 {

Document


src/core/fields/avx512_m31.rs line 70 at r2 (raw file):

    #[inline(always)]
    fn mul(self, rhs: Self) -> Self::Output {

Document

@spapinistarkware spapinistarkware force-pushed the spapini/02-25-Faster_AVX256 branch from 591f7ca to e7624e7 Compare February 26, 2024 13:52
Copy link
Collaborator

@ilyalesokhin-starkware ilyalesokhin-starkware left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 4 files at r2, 2 of 3 files at r3, 1 of 1 files at r4, all commit messages.
Reviewable status: :shipit: complete! all files reviewed, all discussions resolved (waiting on @shaharsamocha7)

@spapinistarkware spapinistarkware merged commit 89ec9e9 into dev Feb 26, 2024
7 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants