Vectorize QuantizeLinear, second stage of DynamicQuantizeLinear #538
Conversation
This file contains 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
robertknight
force-pushed
the
quantize-linear-simd
branch
2 times, most recently
from
1f39c34 to
e7ee69e
Compare
Add SIMD operations that will be useful for implementing vectorized quantization. To enable SIMD operations to return SIMD vectors of narrower data types, needed by `saturating_cast_u8`, a "dummy" impl of `Simd` was added for arrays. This is initially limited to `[u8; N]`.
… ops This is ~3.5x faster with AVX2.
robertknight
force-pushed
the
quantize-linear-simd
branch
from
e7ee69e to
39d98f2
Compare
|
The |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Use SIMD operations for QuantizeLinear in the f32 -> u8 case, which also forms the second stage of DynamicQuantizeLinear. This makes DQL about ~3.5x faster under AVX2.