Add unrolled bit-packing kernels for longs (widths 1..63)#70
Open
raunaqmorarka wants to merge 1 commit into
Open
Add unrolled bit-packing kernels for longs (widths 1..63)#70raunaqmorarka wants to merge 1 commit into
raunaqmorarka wants to merge 1 commit into
Conversation
Branch-free unrolled fastpackwithoutmask/fastunpack for all widths, mirroring the 32-bit BitPacking. Measured on Graviton (aarch64), Corretto 21, median over widths 1..63: pack 1.93x, unpack 2.66x.
raunaqmorarka
force-pushed
the
unrolled-long-bitpacking
branch
from
56c7af8 to
4beeed0
Compare
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Unrolled bit-packing kernels for longs
Add unrolled fastpackwithoutmask{1..63} / fastunpack{1..63} to
LongBitPacking, mirroring the 32-bit BitPacking, and extend both
dispatchers to a 0..64 switch.
Performance
BenchmarkLongBitPacking, Graviton (aarch64), Corretto 21. Speedup of the
unrolled kernels over the generic path: