Micro-optimize performance methods #52555
Closed
+171
−84
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
…correctly Summary: Changelog: [internal] Making `performance.mark` and `performance.measure` slightly more spec-compliant. Differential Revision: D78013571
Summary: Changelog: [internal] Small refactors for `$ReadOnly`, move `getCurrentTimeStamp` to utils so it can be reused, etc. Differential Revision: D78013570
…mance.measure Summary: Changelog: [internal] `PerformanceEntry.detail` should be `null` if not defined, not undefined as it is currently. This fixes that. Differential Revision: D78013568
Summary: Changelog: [internal] This adds a new benchmark for `performance.measure` to measure the performance of calling it without arguments. Differential Revision: D78013563
Summary: Changelog: [internal] Applied several micro-optimizations to methods in the `performance` object: - Cached property access - Avoided unnecessary function calls - Avoided unnecessary object allocations - Use strict comparisons with `undefined` instead of loose comparisons with `null` whenever possible. This yields wins from 5 to 15% depending on the method and args. ## Before | (index) | Task name | Latency average (ns) | Latency median (ns) | Throughput average (ops/s) | Throughput median (ops/s) | Samples | | ------- | --------------------------------------------------------- | -------------------- | ------------------- | -------------------------- | ------------------------- | ------- | | 0 | 'mark (default)' | '1728.19 ± 0.81%' | '1682.00' | '592249 ± 0.01%' | '594530' | 578640 | | 1 | 'mark (with custom startTime)' | '2008.33 ± 1.32%' | '1933.00' | '514326 ± 0.01%' | '517331' | 497927 | | 2 | 'measure (default)' | '2568.90 ± 1.48%' | '2484.00' | '400362 ± 0.02%' | '402576' | 389272 | | 3 | 'measure (with start and end timestamps)' | '2806.21 ± 0.69%' | '2744.00' | '362439 ± 0.02%' | '364431' | 356353 | | 4 | 'measure (with mark names)' | '2860.82 ± 0.23%' | '2824.00' | '352145 ± 0.02%' | '354108' | 349551 | | 5 | 'clearMarks' | '763.05 ± 0.03%' | '751.00' | '1320377 ± 0.01%' | '1331558' | 1310524 | | 6 | 'clearMeasures' | '788.10 ± 0.05%' | '781.00' | '1281839 ± 0.01%' | '1280410' | 1268875 | | 7 | 'mark + clearMarks' | '2129.03 ± 1.04%' | '2053.00' | '484778 ± 0.01%' | '487092' | 469698 | | 8 | 'measure + clearMeasures (with start and end timestamps)' | '3176.38 ± 0.67%' | '3105.00' | '320111 ± 0.02%' | '322061' | 314825 | | 9 | 'measure + clearMeasures (with mark names)' | '3229.33 ± 0.60%' | '3145.00' | '316463 ± 0.02%' | '317965' | 309662 | ## After | (index) | Task name | Latency average (ns) | Latency median (ns) | Throughput average (ops/s) | Throughput median (ops/s) | Samples | | ------- | --------------------------------------------------------- | -------------------- | ------------------- | -------------------------- | ------------------------- | ------- | | 0 | 'mark (default)' | '1598.54 ± 0.93%' | '1553.00' | '639076 ± 0.01%' | '643915' | 625572 | | 1 | 'mark (with custom startTime)' | '1772.07 ± 1.24%' | '1713.00' | '579523 ± 0.01%' | '583771' | 564314 | | 2 | 'measure (default)' | '2349.32 ± 0.81%' | '2284.00' | '434149 ± 0.01%' | '437828' | 425656 | | 3 | 'measure (with start and end timestamps)' | '2609.11 ± 1.06%' | '2534.00' | '391950 ± 0.02%' | '394633' | 383273 | | 4 | 'measure (with mark names)' | '2656.46 ± 0.42%' | '2614.00' | '380128 ± 0.01%' | '382555' | 376442 | | 5 | 'clearMarks' | '666.38 ± 0.03%' | '661.00' | '1511649 ± 0.01%' | '1512859' | 1500641 | | 6 | 'clearMeasures' | '718.62 ± 0.03%' | '711.00' | '1401526 ± 0.01%' | '1406470' | 1391563 | | 7 | 'mark + clearMarks' | '2027.47 ± 1.29%' | '1953.00' | '509105 ± 0.01%' | '512033' | 493227 | | 8 | 'measure + clearMeasures (with start and end timestamps)' | '3030.83 ± 0.98%' | '2915.00' | '340624 ± 0.02%' | '343053' | 329943 | | 9 | 'measure + clearMeasures (with mark names)' | '2915.76 ± 0.59%' | '2844.00' | '349120 ± 0.02%' | '351617' | 342964 | Differential Revision: D78013565
facebook-github-bot
added
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D78013565 |
|
This pull request has been merged in f5abce0. |
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.
Summary:
Changelog: [internal]
Applied several micro-optimizations to methods in the
performanceobject:undefinedinstead of loose comparisons withnullwhenever possible.This yields wins from 5 to 15% depending on the method and args.
Before
After
Differential Revision: D78013565