IH-553: Optimize Sexpr.{escape,unescape} #5576
Merged
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
edwintorok
changed the title
psafont
approved these changes
Apr 19, 2024
|
Can we target master branch here? :) |
|
I targeted perf because I wanted to do more testing, although there is not much that can go wrong here, so I can retarget. |
It was allocating a string for each character, only to then immediately add it to a buffer. Instead add the character to the buffer directly. The temporary string was probably created to factor out 'Buffer.add', but I'd rather copy Buffer.add 4x times than have the perf hit. Further optimizations might be possible here as the comments says, but this is a low-risk, obvious optimization. Signed-off-by: Edwin Török <edwin.torok@cloud.com>
Optimistically check for the presence of escape characters before escaping. This avoids rebuilding the string when escapable characters are not present (the common case). Signed-off-by: Edwin Török <edwin.torok@cloud.com>
Check for the presence of the escape character before rebuilding the string Signed-off-by: Edwin Török <edwin.torok@cloud.com>
edwintorok
force-pushed
the
private/edvint/sexpr
branch
from
ab4e8ce
to
6f15c30
Compare
minglumlu
reviewed
Apr 22, 2024
| ) | ||
| s ; | ||
| Buffer.contents escaped | ||
| Buffer.add_char escaped c |
There was a problem hiding this comment.
Just a question: would a pipe operator be of the same effect (no temporary variable)?
E.g.
(match c with ..) |> Buffer.add_string escaped
minglumlu
approved these changes
Apr 22, 2024
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Outdated suggestions cannot be applied.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.
It was allocating a string for each character, only to then immediately add it to a buffer. Instead add the character to the buffer directly.
The temporary string was probably created to factor out 'Buffer.add', but I'd rather copy Buffer.add 4x times than have the perf hit.
Further optimizations might be possible here as the comments says, but this is a low-risk, obvious optimization.
Before (see how Buffer.add_string and caml_alloc_string take up quite a bit of time):
After: (Buffer.add_string completely gone, no more alloc_string, and the Sexpr.escape is narrower)
In fact we can take this a bit further and check for the presence of escapable characters when escaping and for the presence of the escape character when unescaping. Notice how Sexpr.escape is a lot narrower now, and Sexpr.unescape has disappeared from the flamegraph:
