Arm: Add SVE implementations of simdFilterHor #623
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
micro-arm
force-pushed
the
interpolate-filterhor-sve
branch
from
7b02b70 to
a0ce280
Compare
Contributor
Author
|
Just forced-pushed a fix for a compiler error found in the SVE. I guess none of the CI can compile the SVE paths yet? |
micro-arm
force-pushed
the
interpolate-filterhor-sve
branch
from
a0ce280 to
c8583f8
Compare
The M1 paths for the simdFilter_neon currently use the scalar code for the general filter case. Refactoring it to specialize for case `width == 1` improves the average performance by 2.0x across all taps compared to the current scalar implementation.
Add SVE implementations of the `simdInterpolateHor` filters for the 4-tap, 6-tap, and 8-tap. Performance uplift vs Neon: N8: 1.36x N6: 1.09x N4: 1.09x These benchmarks are obtained from a Neoverse V2 using LLVM 21.
micro-arm
force-pushed
the
interpolate-filterhor-sve
branch
from
c8583f8 to
b552c23
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Arm: Add SVE implementations of simdFilterHor
Add SVE implementations of the simdInterpolateHor filters for the 4-tap, 6-tap, and 8-tap.
Performance uplift vs Neon:
N8: 1.36x
N6: 1.09x
N4: 1.09x
These benchmarks are obtained from a Neoverse V2 using LLVM 21.
Arm: Refactor simdFilter_neon for width = 1 path
width == 1improves the average performance by 2.0x across all taps compared to the current scalar implementation.