Use string create if net6.0 or greater #4183

tonyredondo · 2023-05-25T11:02:33Z

Summary of changes

This PR uses the same approach from dotnet/runtime#86685 when generating short strings.

tracer/src/Datadog.Trace/ClrProfiler/ScopeFactory.cs

tracer/src/Datadog.Trace/Util/UriHelpers.cs

…y/use-string-create

pierotibou · 2023-06-13T10:20:48Z

QQ, as we don't have benchmarks on master, do you confirm it is actually more performant with this code (it's not just .NET 6 that is better?)

tonyredondo · 2023-06-14T16:45:58Z

QQ, as we don't have benchmarks on master, do you confirm it is actually more performant with this code (it's not just .NET 6 that is better?)

Cannot confirm that, because they are really close and could be just noise. Also I don't see any difference in allocations.

datadog-ddstaging · 2023-06-18T13:56:19Z

Datadog Report

Branch report: tony/use-string-create
Commit report: 66f17ab

✅ dd-trace-dotnet: 0 Failed, 0 New Flaky, 294122 Passed, 949 Skipped, 34m 31.72s Wall Time

andrewlock · 2023-06-18T14:18:50Z

Execution-Time Benchmarks Report ⏱️

Execution-time results for samples comparing the following branches/commits:

Execution-time benchmarks measure the whole time it takes to execute a program. And are intended to measure the one-off costs. Cases where the execution time results for the PR are worse than latest master results are shown in red. The following thresholds were used for comparing the execution times:

Welch test with statistical test for significance of 5%
Only results indicating a difference greater than 5% and 5 ms are considered.

Note that these results are based on a single point-in-time result for each branch. For full results, see the dashboard.

Graphs show the p99 interval based on the mean and StdDev of the test run, as well as the mean value of the run (shown as a diamond below the graph).

gantt
    title Execution time (ms) FakeDbCommand (.NET Framework 4.6.2) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4183) - mean (3,021ms)  : 2924, 3118
     .   : milestone, 3021,
    master - mean (3,008ms)  : 2920, 3096
     .   : milestone, 3008,

    section CallTarget+Inlining+NGEN
    This PR (4183) - mean (3,799ms)  : 3709, 3888
     .   : milestone, 3799,
    master - mean (3,814ms)  : 3726, 3901
     .   : milestone, 3814,

gantt
    title Execution time (ms) FakeDbCommand (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4183) - mean (3,117ms)  : 3009, 3225
     .   : milestone, 3117,
    master - mean (3,118ms)  : 3014, 3221
     .   : milestone, 3118,

    section CallTarget+Inlining+NGEN
    This PR (4183) - mean (3,638ms)  : 3557, 3719
     .   : milestone, 3638,
    master - mean (3,624ms)  : 3570, 3678
     .   : milestone, 3624,

gantt
    title Execution time (ms) FakeDbCommand (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4183) - mean (3,110ms)  : 3009, 3210
     .   : milestone, 3110,
    master - mean (3,095ms)  : 3001, 3189
     .   : milestone, 3095,

    section CallTarget+Inlining+NGEN
    This PR (4183) - mean (3,587ms)  : 3530, 3645
     .   : milestone, 3587,
    master - mean (3,629ms)  : 3521, 3738
     .   : milestone, 3629,

gantt
    title Execution time (ms) HttpMessageHandler (.NET Framework 4.6.2) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4183) - mean (189ms)  : 185, 193
     .   : milestone, 189,
    master - mean (190ms)  : 185, 194
     .   : milestone, 190,

    section CallTarget+Inlining+NGEN
    This PR (4183) - mean (1,068ms)  : 1013, 1123
     .   : milestone, 1068,
    master - mean (1,070ms)  : 997, 1143
     .   : milestone, 1070,

gantt
    title Execution time (ms) HttpMessageHandler (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4183) - mean (369ms)  : 363, 374
     .   : milestone, 369,
    master - mean (372ms)  : 365, 379
     .   : milestone, 372,

    section CallTarget+Inlining+NGEN
    This PR (4183) - mean (1,157ms)  : 1131, 1184
     .   : milestone, 1157,
    master - mean (1,157ms)  : 1126, 1189
     .   : milestone, 1157,

gantt
    title Execution time (ms) HttpMessageHandler (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4183) - mean (358ms)  : 352, 365
     .   : milestone, 358,
    master - mean (360ms)  : 352, 368
     .   : milestone, 360,

    section CallTarget+Inlining+NGEN
    This PR (4183) - mean (1,118ms)  : 1085, 1151
     .   : milestone, 1118,
    master - mean (1,119ms)  : 1090, 1148
     .   : milestone, 1119,

andrewlock · 2023-06-18T14:23:20Z

Benchmarks Report 🐌

Benchmarks for #4183 compared to master:

2 benchmarks are faster, with geometric mean 1.164
1 benchmarks have fewer allocations

The following thresholds were used for comparing the benchmark speeds:

Mann–Whitney U test with statistical test for significance of 5%
Only results indicating a difference greater than 10% and 0.3 ns are considered.

Allocation changes below 0.5% are ignored.

Benchmark details

Benchmarks.Trace.AgentWriterBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	624μs	79.8ns	288ns	0	2.62 KB
master	`WriteAndFlushEnrichedTraces`	net472	805μs	278ns	1μs	0.403	3.22 KB
#4183	`WriteAndFlushEnrichedTraces`	net6.0	473μs	338ns	1.31μs	0	2.62 KB
#4183	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	638μs	134ns	518ns	0	2.63 KB
#4183	`WriteAndFlushEnrichedTraces`	net472	804μs	569ns	2.13μs	0.408	3.22 KB

Benchmarks.Trace.AppSecBodyBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`AllCycleSimpleBody`	netcoreapp3.1	40.9μs	71.3ns	276ns	0	0	1.63 KB
master	`AllCycleSimpleBody`	net472	42.4μs	42.1ns	163ns	0.251	0	1.69 KB
master	`AllCycleMoreComplexBody`	netcoreapp3.1	231μs	383ns	1.48μs	0.115	0	9.12 KB
master	`AllCycleMoreComplexBody`	net472	238μs	900ns	4.03μs	1.42	0	9.28 KB
master	`ObjectExtractorSimpleBody`	netcoreapp3.1	183ns	0.299ns	1.16ns	0.00369	0	272 B
master	`ObjectExtractorSimpleBody`	net472	144ns	0.186ns	0.72ns	0.0446	0	281 B
master	`ObjectExtractorMoreComplexBody`	netcoreapp3.1	4.05μs	2.07ns	7.46ns	0.0507	0	3.78 KB
master	`ObjectExtractorMoreComplexBody`	net472	4.14μs	2.06ns	7.41ns	0.617	0.00621	3.89 KB
#4183	`AllCycleSimpleBody`	net6.0	39.2μs	19.9ns	77.2ns	0.019	0	1.65 KB
#4183	`AllCycleSimpleBody`	netcoreapp3.1	40.6μs	103ns	385ns	0.0203	0	1.63 KB
#4183	`AllCycleSimpleBody`	net472	41μs	162ns	627ns	0.264	0	1.69 KB
#4183	`AllCycleMoreComplexBody`	net6.0	219μs	50.9ns	197ns	0.109	0	9.22 KB
#4183	`AllCycleMoreComplexBody`	netcoreapp3.1	228μs	361ns	1.4μs	0.113	0	9.12 KB
#4183	`AllCycleMoreComplexBody`	net472	235μs	36.5ns	132ns	1.41	0	9.28 KB
#4183	`ObjectExtractorSimpleBody`	net6.0	119ns	0.0285ns	0.107ns	0.00391	0	280 B
#4183	`ObjectExtractorSimpleBody`	netcoreapp3.1	178ns	0.0725ns	0.271ns	0.00377	0	272 B
#4183	`ObjectExtractorSimpleBody`	net472	144ns	0.153ns	0.593ns	0.0446	0	281 B
#4183	`ObjectExtractorMoreComplexBody`	net6.0	3.02μs	0.772ns	2.89ns	0.0544	0	3.88 KB
#4183	`ObjectExtractorMoreComplexBody`	netcoreapp3.1	3.95μs	1.38ns	4.98ns	0.0513	0	3.78 KB
#4183	`ObjectExtractorMoreComplexBody`	net472	4.09μs	1.81ns	6.52ns	0.618	0.00614	3.89 KB

Benchmarks.Trace.AspNetCoreBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendRequest`	netcoreapp3.1	188μs	150ns	579ns	0.187	20.08 KB
master	`SendRequest`	net472	0.000965ns	0.000367ns	0.00137ns	0	0 b
#4183	`SendRequest`	net6.0	165μs	80.7ns	313ns	0.246	18.02 KB
#4183	`SendRequest`	netcoreapp3.1	186μs	286ns	1.11μs	0.185	20.08 KB
#4183	`SendRequest`	net472	0.000444ns	0.000225ns	0.000842ns	0	0 b

Benchmarks.Trace.CIVisibilityProtocolWriterBenchmark - Same speed ✔️ Fewer allocations 🎉

Fewer allocations 🎉 in #4183

Benchmark	Base Allocated	Diff Allocated	Change	Change %
Benchmarks.Trace.CIVisibilityProtocolWriterBenchmark.WriteAndFlushEnrichedTraces‑netcoreapp3.1	41.79 KB	41.57 KB	-220 B	-0.53%

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Gen 2	Allocated
master	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	632μs	348ns	1.26μs	0.314	0	0	41.79 KB
master	`WriteAndFlushEnrichedTraces`	net472	798μs	1.59μs	5.94μs	8.2	2.34	0.391	53.23 KB
#4183	`WriteAndFlushEnrichedTraces`	net6.0	500μs	1.59μs	6.15μs	0.496	0	0	41.83 KB
#4183	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	633μs	1.77μs	6.85μs	0.309	0	0	41.57 KB
#4183	`WriteAndFlushEnrichedTraces`	net472	781μs	3.6μs	13.9μs	8.31	2.37	0.396	53.23 KB

Benchmarks.Trace.DbCommandBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`ExecuteNonQuery`	netcoreapp3.1	1.08μs	0.298ns	1.07ns	0.00974	0	720 B
master	`ExecuteNonQuery`	net472	1.31μs	1.21ns	4.68ns	0.108	0	682 B
#4183	`ExecuteNonQuery`	net6.0	874ns	0.257ns	0.994ns	0.0102	0	720 B
#4183	`ExecuteNonQuery`	netcoreapp3.1	1.06μs	0.182ns	0.631ns	0.00968	0	720 B
#4183	`ExecuteNonQuery`	net472	1.42μs	0.46ns	1.59ns	0.108	0.000708	682 B

Benchmarks.Trace.ElasticsearchBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`CallElasticsearch`	netcoreapp3.1	1.28μs	0.579ns	2.09ns	0.0127	0	944 B
master	`CallElasticsearch`	net472	1.92μs	0.484ns	1.87ns	0.151	0.000963	955 B
master	`CallElasticsearchAsync`	netcoreapp3.1	1.31μs	0.676ns	2.53ns	0.013	0	992 B
master	`CallElasticsearchAsync`	net472	2.16μs	0.896ns	3.47ns	0.16	0.00108	1.01 KB
#4183	`CallElasticsearch`	net6.0	850ns	0.231ns	0.863ns	0.0131	0	944 B
#4183	`CallElasticsearch`	netcoreapp3.1	1.22μs	0.276ns	0.996ns	0.0128	0	944 B
#4183	`CallElasticsearch`	net472	2.08μs	0.493ns	1.84ns	0.152	0.00104	955 B
#4183	`CallElasticsearchAsync`	net6.0	887ns	0.264ns	0.989ns	0.0129	0	920 B
#4183	`CallElasticsearchAsync`	netcoreapp3.1	1.25μs	0.483ns	1.67ns	0.0134	0	992 B
#4183	`CallElasticsearchAsync`	net472	2.08μs	0.576ns	2.23ns	0.16	0.00104	1.01 KB

Benchmarks.Trace.GraphQLBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`ExecuteAsync`	netcoreapp3.1	1.32μs	0.365ns	1.37ns	0.012	0	912 B
master	`ExecuteAsync`	net472	1.49μs	0.575ns	2.07ns	0.138	0.000744	875 B
#4183	`ExecuteAsync`	net6.0	1.06μs	0.53ns	1.98ns	0.0128	0	912 B
#4183	`ExecuteAsync`	netcoreapp3.1	1.3μs	1.61ns	6.22ns	0.0122	0	912 B
#4183	`ExecuteAsync`	net472	1.54μs	0.518ns	1.94ns	0.139	0.000772	875 B

Benchmarks.Trace.HttpClientBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendAsync`	netcoreapp3.1	4.3μs	2.08ns	8.04ns	0.0323	2.42 KB
master	`SendAsync`	net472	6.83μs	1.73ns	6.69ns	0.474	2.99 KB
#4183	`SendAsync`	net6.0	3.34μs	0.595ns	2.14ns	0.0264	1.89 KB
#4183	`SendAsync`	netcoreapp3.1	4.13μs	1.13ns	4.36ns	0.0331	2.42 KB
#4183	`SendAsync`	net472	6.79μs	2.42ns	9.38ns	0.474	2.99 KB

Benchmarks.Trace.ILoggerBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	netcoreapp3.1	1.8μs	1.02ns	3.93ns	0.0226	1.62 KB
master	`EnrichedLog`	net472	2.22μs	0.836ns	3.24ns	0.244	1.54 KB
#4183	`EnrichedLog`	net6.0	1.27μs	0.457ns	1.71ns	0.0229	1.62 KB
#4183	`EnrichedLog`	netcoreapp3.1	1.82μs	2.51ns	9.73ns	0.0217	1.62 KB
#4183	`EnrichedLog`	net472	2.22μs	1.6ns	6.21ns	0.244	1.54 KB

Benchmarks.Trace.Log4netBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`EnrichedLog`	netcoreapp3.1	118μs	121ns	468ns	0.0584	0	4.21 KB
master	`EnrichedLog`	net472	148μs	90.4ns	350ns	0.665	0.222	4.38 KB
#4183	`EnrichedLog`	net6.0	109μs	121ns	471ns	0.0544	0	4.21 KB
#4183	`EnrichedLog`	netcoreapp3.1	117μs	256ns	957ns	0.0575	0	4.21 KB
#4183	`EnrichedLog`	net472	147μs	153ns	591ns	0.651	0.217	4.38 KB

Benchmarks.Trace.NLogBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`EnrichedLog`	netcoreapp3.1	3.69μs	1.07ns	3.87ns	0.05	0	3.69 KB
master	`EnrichedLog`	net472	4.79μs	11.5ns	44.5ns	0.526	0.00237	3.31 KB
#4183	`EnrichedLog`	net6.0	3.08μs	0.85ns	3.07ns	0.0522	0	3.71 KB
#4183	`EnrichedLog`	netcoreapp3.1	3.67μs	1.29ns	4.65ns	0.0514	0	3.69 KB
#4183	`EnrichedLog`	net472	4.7μs	1.15ns	4.31ns	0.524	0.00235	3.31 KB

Benchmarks.Trace.RedisBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendReceive`	netcoreapp3.1	1.38μs	0.288ns	1.04ns	0.0131	1 KB
master	`SendReceive`	net472	1.68μs	1.59ns	6.17ns	0.159	1 KB
#4183	`SendReceive`	net6.0	1.07μs	0.439ns	1.64ns	0.0138	1 KB
#4183	`SendReceive`	netcoreapp3.1	1.44μs	0.294ns	1.1ns	0.0137	1 KB
#4183	`SendReceive`	net472	1.73μs	0.719ns	2.49ns	0.159	1 KB

Benchmarks.Trace.SerilogBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	netcoreapp3.1	3.58μs	1.6ns	6ns	0.0198	1.58 KB
master	`EnrichedLog`	net472	4.01μs	4.76ns	18.4ns	0.31	1.96 KB
#4183	`EnrichedLog`	net6.0	2.42μs	0.706ns	2.55ns	0.0212	1.53 KB
#4183	`EnrichedLog`	netcoreapp3.1	3.42μs	0.791ns	2.85ns	0.0207	1.58 KB
#4183	`EnrichedLog`	net472	3.95μs	2.08ns	8.06ns	0.31	1.96 KB

Benchmarks.Trace.SpanBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #4183

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.SpanBenchmark.StartFinishScope‑net472	1.143	847.84	741.58

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`StartFinishSpan`	netcoreapp3.1	475ns	0.639ns	2.47ns	0.00733	536 B
master	`StartFinishSpan`	net472	635ns	0.233ns	0.871ns	0.0854	538 B
master	`StartFinishScope`	netcoreapp3.1	646ns	0.257ns	0.962ns	0.00875	656 B
master	`StartFinishScope`	net472	848ns	0.296ns	1.07ns	0.0982	618 B
#4183	`StartFinishSpan`	net6.0	357ns	0.101ns	0.377ns	0.00762	536 B
#4183	`StartFinishSpan`	netcoreapp3.1	505ns	0.307ns	1.19ns	0.00725	536 B
#4183	`StartFinishSpan`	net472	630ns	0.324ns	1.25ns	0.0852	538 B
#4183	`StartFinishScope`	net6.0	471ns	0.121ns	0.453ns	0.00912	656 B
#4183	`StartFinishScope`	netcoreapp3.1	703ns	1.28ns	4.8ns	0.00898	656 B
#4183	`StartFinishScope`	net472	741ns	0.235ns	0.911ns	0.098	618 B

Benchmarks.Trace.TraceAnnotationsBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #4183

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.TraceAnnotationsBenchmark.RunOnMethodBegin‑netcoreapp3.1	1.184	768.12	648.60

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`RunOnMethodBegin`	netcoreapp3.1	768ns	0.305ns	1.18ns	0.00886	656 B
master	`RunOnMethodBegin`	net472	1.01μs	0.191ns	0.74ns	0.0978	618 B
#4183	`RunOnMethodBegin`	net6.0	578ns	0.26ns	1.01ns	0.00921	656 B
#4183	`RunOnMethodBegin`	netcoreapp3.1	648ns	0.228ns	0.789ns	0.00878	656 B
#4183	`RunOnMethodBegin`	net472	928ns	0.237ns	0.886ns	0.0978	618 B

andrewlock · 2023-06-18T15:27:56Z

Throughput/Crank Report:zap:

Throughput results for AspNetCoreSimpleController comparing the following branches/commits:

Cases where throughput results for the PR are worse than latest master (5% drop or greater), results are shown in red.

Note that these results are based on a single point-in-time result for each branch. For full results, see one of the many, many dashboards!

gantt
    title Throughput Linux x64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4183) (11.282M)   : 0, 11282157
    master (11.444M)   : 0, 11443637
    benchmarks/2.31.0 (11.331M)   : 0, 11330560
    benchmarks/2.9.0 (11.301M)   : 0, 11301309

    section Automatic
    This PR (4183) (8.075M)   : 0, 8074884
    master (8.028M)   : 0, 8027516
    benchmarks/2.31.0 (8.087M)   : 0, 8086789
    benchmarks/2.9.0 (8.193M)   : 0, 8192771

    section Trace stats
    master (8.033M)   : 0, 8032955
    benchmarks/2.31.0 (8.031M)   : 0, 8031178

    section Manual
    This PR (4183) (10.073M)   : 0, 10073433
    master (10.296M)   : 0, 10296400
    benchmarks/2.31.0 (10.125M)   : 0, 10125369

    section Manual + Automatic
    This PR (4183) (7.607M)   : 0, 7606635
    master (7.775M)   : 0, 7775481
    benchmarks/2.31.0 (7.775M)   : 0, 7775145

    section Version Conflict
    master (6.991M)   : 0, 6991290
    benchmarks/2.31.0 (6.977M)   : 0, 6977309

gantt
    title Throughput Linux arm64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4183) (9.562M)   : 0, 9562043
    master (9.492M)   : 0, 9491825
    benchmarks/2.31.0 (9.462M)   : 0, 9461591
    benchmarks/2.9.0 (9.457M)   : 0, 9457357

    section Automatic
    This PR (4183) (6.673M)   : 0, 6672610
    master (6.984M)   : 0, 6983612
    benchmarks/2.31.0 (6.609M)   : 0, 6608731

    section Trace stats
    master (6.809M)   : 0, 6808981
    benchmarks/2.31.0 (6.751M)   : 0, 6750564

    section Manual
    This PR (4183) (8.494M)   : 0, 8493642
    master (8.392M)   : 0, 8392081
    benchmarks/2.31.0 (8.632M)   : 0, 8631992

    section Manual + Automatic
    This PR (4183) (6.573M)   : 0, 6572611
    master (6.690M)   : 0, 6689904
    benchmarks/2.31.0 (6.536M)   : 0, 6535693

    section Version Conflict
    master (5.873M)   : 0, 5872703
    benchmarks/2.31.0 (6.058M)   : 0, 6057526

gantt
    title Throughput Windows x64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4183) (9.561M)   : 0, 9561373
    master (10.487M)   : 0, 10486575
    benchmarks/2.9.0 (9.422M)   : 0, 9421865

    section Automatic
    This PR (4183) (6.923M)   : crit ,0, 6923189
    master (7.547M)   : 0, 7546661
    benchmarks/2.9.0 (7.462M)   : 0, 7461655

    section Trace stats
    master (7.488M)   : 0, 7488208

    section Manual
    This PR (4183) (8.639M)   : crit ,0, 8638501
    master (9.435M)   : 0, 9435451

    section Manual + Automatic
    This PR (4183) (6.699M)   : crit ,0, 6698663
    master (7.296M)   : 0, 7295579

    section Version Conflict
    master (6.688M)   : 0, 6688167

gantt
    title Throughput Linux x64 (ASM) (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4183) (7.263M)   : 0, 7263333
    master (7.248M)   : 0, 7248222
    benchmarks/2.31.0 (7.243M)   : 0, 7243146
    benchmarks/2.9.0 (7.622M)   : 0, 7622034

    section No attack
    This PR (4183) (2.094M)   : 0, 2094241
    master (2.051M)   : 0, 2051432
    benchmarks/2.31.0 (2.364M)   : 0, 2363993
    benchmarks/2.9.0 (3.164M)   : 0, 3163666

    section Attack
    This PR (4183) (1.799M)   : 0, 1798820
    master (1.804M)   : 0, 1804389
    benchmarks/2.31.0 (2.004M)   : 0, 2004139
    benchmarks/2.9.0 (2.518M)   : 0, 2517779

    section Blocking
    This PR (4183) (3.427M)   : 0, 3426620
    master (3.442M)   : 0, 3441577
    benchmarks/2.31.0 (3.967M)   : 0, 3967236

tonyredondo · 2023-06-19T14:04:08Z

QQ, as we don't have benchmarks on master, do you confirm it is actually more performant with this code (it's not just .NET 6 that is better?)

Cannot confirm that, because they are really close and could be just noise. Also I don't see any difference in allocations.

Actually benchmarks are showing some improvements.

andrewlock

I take your word for it on the benchmarks. This string.Create overload is awesome.

andrewlock · 2023-06-19T14:15:47Z

tracer/src/Datadog.Trace/ServiceFabric/ServiceRemotingHelpers.cs

@@ -90,7 +90,11 @@ public static void GetMessageHeaders(EventArgs? eventArgs, out IServiceRemotingR

        public static string GetSpanName(string spanKind)
        {
+#if NET6_0_OR_GREATER
+            return string.Create(null, stackalloc char[128], $"{SpanNamePrefix}.{spanKind}");


Technically there's a potential issue here if spanKind.Length > 128 - SpanNamePrefix.Length. It's very unlikely, and this is internal and we control the spanKind types in this callpath. But it still gives me slight eye twitch! 😄

Use string create if net6.0 or greater

90841ee

tonyredondo self-assigned this May 25, 2023

tonyredondo added 2 commits May 26, 2023 15:31

run benchmarks on net6.0

e326991

Merge branch 'master' into tony/use-string-create

5237deb

CodeBlanch reviewed Jun 7, 2023

View reviewed changes

tracer/src/Datadog.Trace/ClrProfiler/ScopeFactory.cs Outdated Show resolved Hide resolved

CodeBlanch reviewed Jun 7, 2023

View reviewed changes

tracer/src/Datadog.Trace/Util/UriHelpers.cs Outdated Show resolved Hide resolved

tonyredondo added 9 commits June 9, 2023 09:54

Revert some changes

2081f2a

revert

7608a8a

Merge branch 'master' into tony/use-string-create

4af57a2

Fix BenchmarkComparer

7d5e025

Merge remote-tracking branch 'origin/tony/use-string-create' into ton…

08a8382

…y/use-string-create

Merge branch 'master' into tony/use-string-create

b90638f

Ensure both data set exists for comparison.

4ad2dcb

Merge branch 'master' into tony/use-string-create

c268b0c

Merge branch 'master' into tony/use-string-create

bd7a80a

tonyredondo marked this pull request as ready for review June 12, 2023 21:08

tonyredondo requested review from a team as code owners June 12, 2023 21:08

Merge branch 'master' into tony/use-string-create

648e659

DataDog deleted a comment from datadog-ddstaging bot Jun 14, 2023

DataDog deleted a comment from andrewlock Jun 14, 2023

DataDog deleted a comment from datadog-ddstaging bot Jun 14, 2023

DataDog deleted a comment from andrewlock Jun 14, 2023

Merge branch 'master' into tony/use-string-create

4d15911

tonyredondo added 2 commits June 15, 2023 22:10

Merge branch 'master' into tony/use-string-create

40422d4

Merge branch 'master' into tony/use-string-create

d391d08

DataDog deleted a comment from datadog-ddstaging bot Jun 18, 2023

DataDog deleted a comment from andrewlock Jun 18, 2023

Merge branch 'master' into tony/use-string-create

66f17ab

andrewlock approved these changes Jun 19, 2023

View reviewed changes

pierotibou approved these changes Jun 19, 2023

View reviewed changes

tonyredondo merged commit 89ce11d into master Jun 20, 2023
86 checks passed

tonyredondo deleted the tony/use-string-create branch June 20, 2023 13:43

github-actions bot added this to the vNext milestone Jun 20, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Use string create if net6.0 or greater #4183

Use string create if net6.0 or greater #4183

tonyredondo commented May 25, 2023 •

edited

pierotibou commented Jun 13, 2023

tonyredondo commented Jun 14, 2023

datadog-ddstaging bot commented Jun 18, 2023 •

edited

andrewlock commented Jun 18, 2023

andrewlock commented Jun 18, 2023

Raw results

Raw results

Raw results

Fewer allocations 🎉 in #4183

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Faster 🎉 in #4183

Raw results

Faster 🎉 in #4183

Raw results

andrewlock commented Jun 18, 2023

tonyredondo commented Jun 19, 2023

andrewlock left a comment

andrewlock Jun 19, 2023

Use string create if net6.0 or greater #4183

Use string create if net6.0 or greater #4183

Conversation

tonyredondo commented May 25, 2023 • edited

Summary of changes

pierotibou commented Jun 13, 2023

tonyredondo commented Jun 14, 2023

datadog-ddstaging bot commented Jun 18, 2023 • edited

Datadog Report

andrewlock commented Jun 18, 2023

Execution-Time Benchmarks Report ⏱️

andrewlock commented Jun 18, 2023

Benchmarks Report 🐌

Benchmark details

Raw results

Raw results

Raw results

Fewer allocations 🎉 in #4183

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Faster 🎉 in #4183

Raw results

Faster 🎉 in #4183

Raw results

andrewlock commented Jun 18, 2023

Throughput/Crank Report:zap:

tonyredondo commented Jun 19, 2023

andrewlock left a comment

Choose a reason for hiding this comment

andrewlock Jun 19, 2023

Choose a reason for hiding this comment

tonyredondo commented May 25, 2023 •

edited

datadog-ddstaging bot commented Jun 18, 2023 •

edited