[Perf] Regressions in System.Buffers.Tests.RentReturnArrayPoolTests<Object>

[performanceautofiler]

Run Information

Architecture	arm64
OS	Windows 10.0.19041
Baseline	ec3f9fa03f72ceadfa0610196a12bab3b37b63f9
Compare	775ecaefe7aa3be85858e612fef782f0e1991f51
Diff	Diff

Regressions in System.Buffers.Tests.RentReturnArrayPoolTests<Object>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio	Baseline ETL	Compare ETL
MultipleSerial - Duration of single invocation	814.29 ns	881.75 ns	1.08	0.07	False

Historical Data in Reporting System

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f netcoreapp5.0 --filter 'System.Buffers.Tests.RentReturnArrayPoolTests<Object>*'

Payloads

Baseline
Compare

Histogram

System.Buffers.Tests.RentReturnArrayPoolTests<Object>.MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: True)

Docs

Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository

[dotnet-issue-labeler]

I couldn't figure out the best area label to add to this issue. If you have write-permissions please help me learn by adding exactly one area label.

[DrewScoggins]

Seems related to #55959

[stephentoub]

@DrewScoggins, how do I see other generic instantations of this same test for the same period? The test runs for both object and byte:
https://github.com/dotnet/performance/blob/d7dac8a7ca12a28d099192f8a901cf8e30361384/src/benchmarks/micro/libraries/System.Buffers/ArrayPoolTests.cs#L14-L15

[DrewScoggins]

So the link is here, https://pvscmdupload.blob.core.windows.net/reports/allTestHistory/refs/heads/main_arm64_Windows%2010.0.19041/System.Buffers.Tests.RentReturnArrayPoolTests(Byte).MultipleSerial(RentalSize%3a%204096%2c%20ManipulateArray%3a%20False%2c%20Async%3a%20False%2c%20UseSharedPool%3a%20True).html

I got this by just clicking on the link in the table above and replacing the Object inside parens with Byte.

[ghost]

Tagging subscribers to this area: @GrabYourPitchforks, @dotnet/area-system-buffers
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Run Information

Architecture	arm64
OS	Windows 10.0.19041
Baseline	ec3f9fa03f72ceadfa0610196a12bab3b37b63f9
Compare	775ecaefe7aa3be85858e612fef782f0e1991f51
Diff	Diff

Regressions in System.Buffers.Tests.RentReturnArrayPoolTests<Object>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio	Baseline ETL	Compare ETL
MultipleSerial - Duration of single invocation	814.29 ns	881.75 ns	1.08	0.07	False

Historical Data in Reporting System

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f netcoreapp5.0 --filter 'System.Buffers.Tests.RentReturnArrayPoolTests<Object>*'

Payloads

Baseline
Compare

Histogram

System.Buffers.Tests.RentReturnArrayPoolTests<Object>.MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: True)

Docs

Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository

Author:	performanceautofiler[bot]
Assignees:	-
Labels:	arch-arm64, area-System.Buffers, os-windows, tenet-performance, tenet-performance-benchmarks, untriaged
Milestone:	-

[tannergooding]

@DrewScoggins, @stephentoub, was this addressed by #56316 ?

[stephentoub]

Apparently partially but not fully:
#56316 (comment)

[tannergooding]

A 2.56% regression (20ns) isn't bad. Do we think there are additional gains that can still be won back?

If so, I'll put this in future; otherwise, I think we should probably close the issue and accept the regression (given the wins we otherwise got).

[stephentoub]

I think future (or closed) here is fine. It's a tiny hit in one use case on one architecture, and apparently only for object but not byte/char.