Skip to content

ENH: Add new intrinsics sum_u8/u16/u64.#18200

Merged
mattip merged 10 commits intonumpy:masterfrom
Qiyu8:intrin-sum
Feb 2, 2021
Merged

ENH: Add new intrinsics sum_u8/u16/u64.#18200
mattip merged 10 commits intonumpy:masterfrom
Qiyu8:intrin-sum

Conversation

@Qiyu8
Copy link
Copy Markdown
Member

@Qiyu8 Qiyu8 commented Jan 21, 2021

Inspired by #18183 ,we need to add new intrinsics in order to get a better acceleration.

@charris charris changed the title Add new intrinsics sum_u8/u16/u64. ENH: Add new intrinsics sum_u8/u16/u64. Jan 22, 2021
@charris charris added 03 - Maintenance component: SIMD Issues in SIMD (fast instruction sets) code or machinery and removed 03 - Maintenance labels Jan 22, 2021
Copy link
Copy Markdown
Member

@seiko2plus seiko2plus left a comment

Choose a reason for hiding this comment

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

several fixes and improvements

Copy link
Copy Markdown
Member

@seiko2plus seiko2plus left a comment

Choose a reason for hiding this comment

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

Cleanup & several improvements for SSE

@Qiyu8
Copy link
Copy Markdown
Member Author

Qiyu8 commented Jan 25, 2021

The code change here should not trigger the CI test failure of numpy.core.tests.test_multiarray.TestMethods.test_no_dgemv, will restart for verification.

@Qiyu8 Qiyu8 closed this Jan 25, 2021
@Qiyu8 Qiyu8 reopened this Jan 25, 2021
@Qiyu8 Qiyu8 requested a review from seiko2plus January 25, 2021 09:23
@touqir14
Copy link
Copy Markdown
Contributor

touqir14 commented Feb 1, 2021

Any updates here @seiko2plus @Qiyu8 ? I can update PR 18183 once this gets merged.

Copy link
Copy Markdown
Member

@seiko2plus seiko2plus left a comment

Choose a reason for hiding this comment

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

LGTM, Thank you!

@mattip mattip merged commit 78cf92c into numpy:master Feb 2, 2021
@mattip
Copy link
Copy Markdown
Member

mattip commented Feb 2, 2021

Thanks @Qiyu8

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

01 - Enhancement component: SIMD Issues in SIMD (fast instruction sets) code or machinery

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants