Assume NativePerformance will be available if the Performance module is loaded #52671
+89
−136
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
…facebook#52668) Summary: Changelog: [internal] These methods have been available for months so no need to consider backwards compatibility with older binaries. Reviewed By: huntie Differential Revision: D78412932
…is loaded Summary: Changelog: [internal] When setting up the global `performance` value, we check if the native module exists to decide what version to setup: the legacy "barebones" version or the modern version that's mostly spec compliant. As we do that check and we shouldn't be requiring the module from anywhere else, it should be safe to always assume the module will be defined if we load the `Performance` class, so we can avoid checks in all methods. NOTE: I've kept some checks for a few methods that are still not fully propagated (new methods like `reportMark`, `reportMeasure` and `getMarkTime`). This improves performance slightly for `performance` methods: * Before | (index) | Task name | Latency average (ns) | Latency median (ns) | Throughput average (ops/s) | Throughput median (ops/s) | Samples | | ------- | --------------------------------------------------------- | -------------------- | ------------------- | -------------------------- | ------------------------- | ------- | | 0 | 'mark (default)' | '1803.32 ± 0.71%' | '1763.00' | '563849 ± 0.01%' | '567215' | 554908 | | 1 | 'mark (with custom startTime)' | '1762.39 ± 1.67%' | '1692.00' | '587832 ± 0.01%' | '591017' | 567414 | | 2 | 'measure (default)' | '1879.47 ± 1.44%' | '1813.00' | '548962 ± 0.01%' | '551572' | 532064 | | 3 | 'measure (with start and end timestamps)' | '1916.65 ± 0.84%' | '1873.00' | '531258 ± 0.01%' | '533903' | 521743 | | 4 | 'measure (with mark names)' | '2049.84 ± 0.44%' | '2013.00' | '492799 ± 0.01%' | '496771' | 487844 | | 5 | 'clearMarks' | '719.87 ± 0.04%' | '711.00' | '1403602 ± 0.01%' | '1406470' | 1389136 | | 6 | 'clearMeasures' | '710.04 ± 0.04%' | '701.00' | '1421610 ± 0.01%' | '1426534' | 1408373 | | 7 | 'mark + clearMarks' | '2256.56 ± 1.10%' | '2143.00' | '460721 ± 0.02%' | '466636' | 443152 | | 8 | 'measure + clearMeasures (with start and end timestamps)' | '2345.44 ± 1.11%' | '2244.00' | '442395 ± 0.02%' | '445633' | 426360 | | 9 | 'measure + clearMeasures (with mark names)' | '2349.55 ± 0.61%' | '2283.00' | '434370 ± 0.02%' | '438020' | 425613 | * After | (index) | Task name | Latency average (ns) | Latency median (ns) | Throughput average (ops/s) | Throughput median (ops/s) | Samples | | ------- | --------------------------------------------------------- | -------------------- | ------------------- | -------------------------- | ------------------------- | ------- | | 0 | 'mark (default)' | '1791.47 ± 1.08%' | '1732.00' | '573787 ± 0.01%' | '577367' | 558202 | | 1 | 'mark (with custom startTime)' | '1699.41 ± 1.27%' | '1642.00' | '605188 ± 0.01%' | '609013' | 588441 | | 2 | 'measure (default)' | '1820.92 ± 1.39%' | '1763.00' | '563437 ± 0.01%' | '567215' | 549173 | | 3 | 'measure (with start and end timestamps)' | '1923.57 ± 1.65%' | '1852.00' | '537112 ± 0.01%' | '539957' | 519867 | | 4 | 'measure (with mark names)' | '2036.09 ± 1.05%' | '1983.00' | '500406 ± 0.01%' | '504286' | 491139 | | 5 | 'clearMarks' | '657.49 ± 0.07%' | '641.00' | '1543793 ± 0.01%' | '1560062' | 1520939 | | 6 | 'clearMeasures' | '669.02 ± 0.09%' | '651.00' | '1520386 ± 0.01%' | '1536098' | 1494730 | | 7 | 'mark + clearMarks' | '2213.09 ± 1.53%' | '2103.00' | '470008 ± 0.02%' | '475511' | 451858 | | 8 | 'measure + clearMeasures (with start and end timestamps)' | '2353.15 ± 1.27%' | '2214.00' | '448563 ± 0.02%' | '451671' | 424962 | | 9 | 'measure + clearMeasures (with mark names)' | '2298.28 ± 0.62%' | '2243.00' | '441703 ± 0.02%' | '445831' | 435108 | Reviewed By: hoxyq Differential Revision: D78412931
meta-cla
bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D78412931 |
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]
When setting up the global
performancevalue, we check if the native module exists to decide what version to setup: the legacy "barebones" version or the modern version that's mostly spec compliant.As we do that check and we shouldn't be requiring the module from anywhere else, it should be safe to always assume the module will be defined if we load the
Performanceclass, so we can avoid checks in all methods.NOTE: I've kept some checks for a few methods that are still not fully propagated (new methods like
reportMark,reportMeasureandgetMarkTime).This improves performance slightly for
performancemethods:Reviewed By: hoxyq
Differential Revision: D78412931