Skip to content

hal: vsx: further optimize v_signmask#15235

Merged
opencv-pushbot merged 1 commit intoopencv:3.4from
pmur:vsx-v_signmask-vbpermq
Aug 6, 2019
Merged

hal: vsx: further optimize v_signmask#15235
opencv-pushbot merged 1 commit intoopencv:3.4from
pmur:vsx-v_signmask-vbpermq

Conversation

@pmur
Copy link
Copy Markdown
Contributor

@pmur pmur commented Aug 5, 2019

Use the quadword bit permutation instruction to creatively move
the sign bits to create the mask. Note that permute values above 127
will result in 0 in the respective bit position.

This results in a 5-15% speedup in the houghlines imgproc benchmarks.
Otherwise, no change is observed.

Use the quadword bit permutation instruction to creatively move
the sign bits to create the mask. Note that values above 127 will
result in 0.
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.

Well done! Thank you 👍

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

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants