ENH: Optimize LUT with transposed loads and 4-way row unrolling #7
+236
−86
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
This patch add several improvements to the Lookup Table (LUT) for non-scalable architectures. - **Transposed loads:** Introduced a pre-calculated transposed storage (`trans_`) and loading strategy. This allows using optimized interleaved loads (X2 and X4) when vector lanes match the blocking factor, reducing the need for expensive gather operations. - **FourTablesLookup:** Handle case where the vector length is exactly one-quarter of the table width using 4-way table lookups. Maintenance: - Removed runtime specializations for scalable extensions. These now explicitly fallback to Highway's `GatherIndex` (no performance gain on SVE).
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
This patch add several improvements to the Lookup Table (LUT) for non-scalable architectures.
trans_) and loading strategy. This allows using optimized interleaved loads (X2 and X4) when vector lanes match the blocking factor, reducing the need for expensive gather operations.Maintenance:
GatherIndex(no performance gain on SVE).