Skip to content

intrin: fixed int64->double conversion for AVX-512#16635

Merged
opencv-pushbot merged 1 commit intoopencv:3.4from
mshabunin:fix-avx512-cvt
Feb 21, 2020
Merged

intrin: fixed int64->double conversion for AVX-512#16635
opencv-pushbot merged 1 commit intoopencv:3.4from
mshabunin:fix-avx512-cvt

Conversation

@mshabunin
Copy link
Copy Markdown
Contributor

@mshabunin mshabunin commented Feb 21, 2020 •
edited by alalek
Loading

related #15510

Wrong intrinsics were used in AVX-512 without DQ fallback implementation.

Pull Request Readiness Checklist

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

  • I agree to contribute to the project under OpenCV (BSD) License.
  • To the best of my knowledge, the proposed patch is not based on a code under GPL or other license that is incompatible with OpenCV
  • The PR is proposed to proper branch
  • There is reference to original bug report and related work
  • There is accuracy test, performance test and test data in opencv_extra repository, if applicable
    Patch to opencv_extra has the same branch name.
  • The feature is well documented and sample code can be built with the project CMake
force_builders=Linux AVX2,Custom
buildworker:Custom=linux-3
build_image:Custom=ubuntu:18.04
CPU_BASELINE:Custom=AVX512_SKX
disable_ipp=ON

seiko2plus
seiko2plus reviewed Feb 21, 2020
View reviewed changes
modules/core/include/opencv2/core/hal/intrin_avx512.hpp
@@ -1553,13 +1553,13 @@ inline v_float64x8 v_cvt_f64(const v_int64x8& v)
return v_float64x8(_mm512_cvtepi64_pd(v.val));
Copy link
Copy Markdown
Contributor

@seiko2plus seiko2plus Feb 21, 2020 •
edited
Loading

Choose a reason for hiding this comment

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

My bad I just forget to remove this branch. however, no need to emulate it, since the minimum support for AVX512 support is SKX features, which always enables AVX512DQ.

alalek
alalek approved these changes Feb 21, 2020
View reviewed changes
Copy link
Copy Markdown
Member

@alalek alalek left a comment

Choose a reason for hiding this comment

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

Thank you!

@opencv-pushbot opencv-pushbot merged commit a0f5eb2 into opencv:3.4 Feb 21, 2020
@mshabunin mshabunin deleted the fix-avx512-cvt branch February 21, 2020 13:43
@alalek alalek mentioned this pull request Feb 26, 2020
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@alalek alalek alalek approved these changes

+1 more reviewer

@seiko2plus seiko2plus seiko2plus left review comments

Reviewers whose approvals may not affect merge requirements

Assignees

@alalek alalek

Labels

category: build/install category: core

Projects

None yet

Milestone

3.4.10

Development

Successfully merging this pull request may close these issues.

4 participants

@mshabunin @alalek @seiko2plus @opencv-pushbot