Remove Boyer-Moore in favor of IndexOf #51815
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
BrennanConroy
added
Perf
area-networking
BrennanConroy
requested review from
Tratcher,
captainsafia and
halter73
as code owners
Tratcher
approved these changes
Nov 1, 2023
gfoidl
reviewed
Nov 2, 2023
ghost
added this to the
1 task
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.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.
Closes #49223
Continuation of #51426
This PR removes boyer-moore entirely in favor of the vectorized
IndexOfmethod. This removes anint[256]alloc per section of the multipart data, as well as the construction cost of generating the skip table.Additionally, noticed the
MultipartBoundaryclass was allocating the boundary bytes array twice per use. The only difference between the arrays is one has\r\nat the beginning, so we can avoid the second alloc by always using the larger array and slicing.Before:
After:
Highlights:
When the boundary size was small (6) and the payload was large (10m+ bytes) the RPS was ~120 and ~140 (no read vs. read), with the change it's now ~970 and ~1300.
Smaller payload tests improved by ~10%-33%.