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.

Sign up for GitHub

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

Jump to bottom

AltiVec gfx/ycbcr has some inefficiencies #470

Closed
classilla opened this issue Jan 27, 2018 · 2 comments
Closed

AltiVec gfx/ycbcr has some inefficiencies #470

classilla opened this issue Jan 27, 2018 · 2 comments

Comments

@classilla
Copy link
Owner

  • Loading from the YUV table is not 16-byte aligned, and we load a LOT of data.
  • The unaligned stores are probably not as efficient as they could be.
classilla added a commit that referenced this issue Jan 27, 2018
@classilla
#470: convert YUV table to AltiVec alignment
7b68aa3
@classilla
Copy link
Owner Author

Actually, the unaligned stores are probably optimal since we have no alignment guarantees in the original x86. Even switching to double could hurt, since the doubles may not be 8-byte aligned, and would cause a hardware fault. But there are still a couple other things to do.

classilla added a commit that referenced this issue Jan 27, 2018
@classilla
#470: remove unproductive optimizations from row filter
8f15c68
classilla added a commit that referenced this issue Jan 27, 2018
@classilla
#470: clean up row filters, collapse into permute
d7cd088
@classilla
Copy link
Owner Author

This is notably better and I'm not sure we can improve further, so let's call it good.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@classilla