Load the tracer/profiler after guardrails checks #5968

kevingosse · 2024-08-30T12:16:40Z

Summary of changes

Instead of loading the native trace/profiler as soon as the native loader is loaded, wait until we've checked the target runtime/process name.

Reason for change

When returning an error code from ICorProfilerCallback::Initialize, .NET unloads the native loader. However, the native tracer/profiler stay in memory (because .NET is not aware of them). Therefore it makes sense to perform as many checks as possible before actually loading them. Plus that's less work to do in the most common failure cases (blocklisted process).

There are still cases where we're not unloading the native tracer/profiler even though we should (for instance, if their Initialize method failed). However they're trickier to get right (we must make sure there's no background thread running, for instance the logger) so I'll leave that for a future PR.

Implementation details

Moved the LoadConfiguration/LoadClassFactory/LoadInstance methods of the dynamic dispatcher into a single Initialize method.

Test coverage

Other details

andrewlock

Kinda taking your word for it here...

shared/src/Datadog.Trace.ClrProfiler.Native/dynamic_dispatcher.cpp

datadog-ddstaging · 2024-08-30T12:46:41Z

Datadog Report

Branch report: kevin/better_loader
Commit report: 07e6afd
Test service: dd-trace-dotnet

✅ 0 Failed, 369378 Passed, 2384 Skipped, 18h 8m 14.47s Total Time
❄️ 1 New Flaky

New Flaky Tests (1)

EarlyFlakeDetection - Datadog.Trace.ClrProfiler.IntegrationTests.CI.PipesXUnitEvpTests - Last Failure
Expand for error
```
 
 Actual:   114 Failure
```

andrewlock · 2024-09-02T08:00:45Z

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 (5968) - mean (70ms)  : 68, 72
     .   : milestone, 70,
    master - mean (70ms)  : 68, 72
     .   : milestone, 70,

    section CallTarget+Inlining+NGEN
    This PR (5968) - mean (1,109ms)  : 1089, 1129
     .   : milestone, 1109,
    master - mean (1,105ms)  : 1087, 1123
     .   : milestone, 1105,

gantt
    title Execution time (ms) FakeDbCommand (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (5968) - mean (109ms)  : 105, 113
     .   : milestone, 109,
    master - mean (109ms)  : 106, 112
     .   : milestone, 109,

    section CallTarget+Inlining+NGEN
    This PR (5968) - mean (775ms)  : 762, 787
     .   : milestone, 775,
    master - mean (772ms)  : 756, 787
     .   : milestone, 772,

gantt
    title Execution time (ms) FakeDbCommand (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (5968) - mean (93ms)  : 90, 95
     .   : milestone, 93,
    master - mean (93ms)  : 90, 96
     .   : milestone, 93,

    section CallTarget+Inlining+NGEN
    This PR (5968) - mean (733ms)  : 713, 753
     .   : milestone, 733,
    master - mean (727ms)  : 709, 744
     .   : milestone, 727,

gantt
    title Execution time (ms) HttpMessageHandler (.NET Framework 4.6.2) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (5968) - mean (190ms)  : 186, 194
     .   : milestone, 190,
    master - mean (191ms)  : 187, 194
     .   : milestone, 191,

    section CallTarget+Inlining+NGEN
    This PR (5968) - mean (1,203ms)  : 1156, 1249
     .   : milestone, 1203,
    master - mean (1,194ms)  : 1165, 1223
     .   : milestone, 1194,

gantt
    title Execution time (ms) HttpMessageHandler (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (5968) - mean (277ms)  : 272, 281
     .   : milestone, 277,
    master - mean (276ms)  : 271, 280
     .   : milestone, 276,

    section CallTarget+Inlining+NGEN
    This PR (5968) - mean (942ms)  : 924, 959
     .   : milestone, 942,
    master - mean (936ms)  : 920, 952
     .   : milestone, 936,

gantt
    title Execution time (ms) HttpMessageHandler (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (5968) - mean (265ms)  : 261, 269
     .   : milestone, 265,
    master - mean (265ms)  : 261, 269
     .   : milestone, 265,

    section CallTarget+Inlining+NGEN
    This PR (5968) - mean (928ms)  : 906, 950
     .   : milestone, 928,
    master - mean (923ms)  : 906, 940
     .   : milestone, 923,

andrewlock · 2024-09-02T08:31:54Z

Benchmarks Report for tracer 🐌

Benchmarks for #5968 compared to master:

1 benchmarks are faster, with geometric mean 1.130
5 benchmarks are slower, with geometric mean 1.147
1 benchmarks have more 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.ActivityBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Gen 2	Allocated
master	`StartStopWithChild`	net6.0	7.6μs	42.4ns	281ns	0.0177	0.00706	0	5.43 KB
master	`StartStopWithChild`	netcoreapp3.1	9.8μs	50.6ns	253ns	0.0198	0.0099	0	5.62 KB
master	`StartStopWithChild`	net472	16μs	58.1ns	225ns	1.03	0.316	0.111	6.06 KB
#5968	`StartStopWithChild`	net6.0	7.7μs	43.8ns	333ns	0.0115	0.00383	0	5.43 KB
#5968	`StartStopWithChild`	netcoreapp3.1	10.1μs	55.8ns	366ns	0.019	0.00952	0	5.62 KB
#5968	`StartStopWithChild`	net472	16.1μs	50.5ns	196ns	1.02	0.319	0.104	6.07 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`WriteAndFlushEnrichedTraces`	net6.0	497μs	327ns	1.18μs	0	2.7 KB
master	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	637μs	319ns	1.15μs	0	2.7 KB
master	`WriteAndFlushEnrichedTraces`	net472	825μs	509ns	1.91μs	0.411	3.3 KB
#5968	`WriteAndFlushEnrichedTraces`	net6.0	486μs	141ns	507ns	0	2.7 KB
#5968	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	632μs	285ns	1.07μs	0	2.7 KB
#5968	`WriteAndFlushEnrichedTraces`	net472	836μs	571ns	2.21μs	0.417	3.3 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendRequest`	net6.0	194μs	1.1μs	8.47μs	0.186	18.45 KB
master	`SendRequest`	netcoreapp3.1	214μs	1.22μs	9.16μs	0.214	20.6 KB
master	`SendRequest`	net472	0.000531ns	0.000327ns	0.00127ns	0	0 b
#5968	`SendRequest`	net6.0	190μs	722ns	2.7μs	0.196	18.45 KB
#5968	`SendRequest`	netcoreapp3.1	213μs	1.19μs	8.06μs	0.205	20.61 KB
#5968	`SendRequest`	net472	0.000427ns	0.000263ns	0.000985ns	0	0 b

Benchmarks.Trace.CIVisibilityProtocolWriterBenchmark - Same speed ✔️ More allocations ⚠️

More allocations ⚠️ in #5968

Benchmark	Base Allocated	Diff Allocated	Change	Change %
Benchmarks.Trace.CIVisibilityProtocolWriterBenchmark.WriteAndFlushEnrichedTraces‑net6.0	41.55 KB	41.84 KB	287 B	0.69%

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Gen 2	Allocated
master	`WriteAndFlushEnrichedTraces`	net6.0	562μs	2.25μs	8.7μs	0.548	0	0	41.55 KB
master	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	676μs	2.61μs	9.4μs	0.338	0	0	41.74 KB
master	`WriteAndFlushEnrichedTraces`	net472	892μs	3.48μs	13μs	8.87	2.66	0.443	53.31 KB
#5968	`WriteAndFlushEnrichedTraces`	net6.0	581μs	2.95μs	12.9μs	0.566	0	0	41.84 KB
#5968	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	665μs	2.59μs	9.67μs	0.327	0	0	41.71 KB
#5968	`WriteAndFlushEnrichedTraces`	net472	880μs	3.96μs	17.7μs	8.36	2.64	0.44	53.26 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`ExecuteNonQuery`	net6.0	1.25μs	0.755ns	2.92ns	0.0143	1.02 KB
master	`ExecuteNonQuery`	netcoreapp3.1	1.81μs	3.88ns	14.5ns	0.0134	1.02 KB
master	`ExecuteNonQuery`	net472	2.04μs	1.54ns	5.74ns	0.157	987 B
#5968	`ExecuteNonQuery`	net6.0	1.23μs	1.92ns	7.45ns	0.0141	1.02 KB
#5968	`ExecuteNonQuery`	netcoreapp3.1	1.72μs	1.06ns	4.1ns	0.0137	1.02 KB
#5968	`ExecuteNonQuery`	net472	2.17μs	3.59ns	13.9ns	0.157	987 B

Benchmarks.Trace.ElasticsearchBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #5968

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.ElasticsearchBenchmark.CallElasticsearchAsync‑net6.0	1.130	1,356.72	1,200.36

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`CallElasticsearch`	net6.0	1.28μs	0.852ns	3.3ns	0.0135	976 B
master	`CallElasticsearch`	netcoreapp3.1	1.54μs	1.66ns	6.44ns	0.0133	976 B
master	`CallElasticsearch`	net472	2.45μs	2.64ns	10.2ns	0.158	995 B
master	`CallElasticsearchAsync`	net6.0	1.36μs	0.814ns	3.05ns	0.0129	952 B
master	`CallElasticsearchAsync`	netcoreapp3.1	1.6μs	0.962ns	3.6ns	0.0136	1.02 KB
master	`CallElasticsearchAsync`	net472	2.51μs	1.34ns	4.84ns	0.166	1.05 KB
#5968	`CallElasticsearch`	net6.0	1.2μs	0.811ns	3.14ns	0.0138	976 B
#5968	`CallElasticsearch`	netcoreapp3.1	1.51μs	0.543ns	1.96ns	0.0134	976 B
#5968	`CallElasticsearch`	net472	2.6μs	1.66ns	6.44ns	0.158	995 B
#5968	`CallElasticsearchAsync`	net6.0	1.2μs	0.698ns	2.61ns	0.0133	952 B
#5968	`CallElasticsearchAsync`	netcoreapp3.1	1.69μs	0.514ns	1.92ns	0.0135	1.02 KB
#5968	`CallElasticsearchAsync`	net472	2.64μs	1.99ns	7.72ns	0.166	1.05 KB

Benchmarks.Trace.GraphQLBenchmark - Slower ⚠️ Same allocations ✔️

Slower ⚠️ in #5968

Benchmark	diff/base	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.GraphQLBenchmark.ExecuteAsync‑net6.0	1.117	1,189.76	1,329.31

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`ExecuteAsync`	net6.0	1.19μs	0.992ns	3.71ns	0.0132	952 B
master	`ExecuteAsync`	netcoreapp3.1	1.59μs	0.213ns	0.737ns	0.0127	952 B
master	`ExecuteAsync`	net472	1.82μs	0.903ns	3.38ns	0.145	915 B
#5968	`ExecuteAsync`	net6.0	1.33μs	1.24ns	4.81ns	0.0133	952 B
#5968	`ExecuteAsync`	netcoreapp3.1	1.6μs	7.31ns	28.3ns	0.0122	952 B
#5968	`ExecuteAsync`	net472	1.84μs	1.23ns	4.77ns	0.145	915 B

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendAsync`	net6.0	4.3μs	1.19ns	4.47ns	0.0302	2.22 KB
master	`SendAsync`	netcoreapp3.1	5.09μs	2.67ns	10.4ns	0.0383	2.76 KB
master	`SendAsync`	net472	7.85μs	14.2ns	55.2ns	0.498	3.15 KB
#5968	`SendAsync`	net6.0	4.27μs	3.16ns	11.8ns	0.0322	2.22 KB
#5968	`SendAsync`	netcoreapp3.1	5.15μs	12ns	46.5ns	0.0362	2.76 KB
#5968	`SendAsync`	net472	7.8μs	1.17ns	4.52ns	0.499	3.15 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net6.0	1.46μs	0.853ns	3.31ns	0.0233	1.64 KB
master	`EnrichedLog`	netcoreapp3.1	2.18μs	0.809ns	3.13ns	0.0223	1.64 KB
master	`EnrichedLog`	net472	2.45μs	0.71ns	2.56ns	0.25	1.57 KB
#5968	`EnrichedLog`	net6.0	1.59μs	0.966ns	3.74ns	0.0232	1.64 KB
#5968	`EnrichedLog`	netcoreapp3.1	2.37μs	1.08ns	3.9ns	0.0212	1.64 KB
#5968	`EnrichedLog`	net472	2.56μs	1.71ns	6.62ns	0.249	1.57 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`EnrichedLog`	net6.0	114μs	118ns	456ns	0.057	0	4.28 KB
master	`EnrichedLog`	netcoreapp3.1	118μs	205ns	794ns	0	0	4.28 KB
master	`EnrichedLog`	net472	150μs	840ns	5.44μs	0.656	0.219	4.46 KB
#5968	`EnrichedLog`	net6.0	115μs	194ns	753ns	0.0573	0	4.28 KB
#5968	`EnrichedLog`	netcoreapp3.1	121μs	139ns	538ns	0	0	4.28 KB
#5968	`EnrichedLog`	net472	146μs	102ns	393ns	0.659	0.22	4.46 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net6.0	2.93μs	0.6ns	2.16ns	0.0306	2.2 KB
master	`EnrichedLog`	netcoreapp3.1	4.26μs	1.91ns	7.39ns	0.0298	2.2 KB
master	`EnrichedLog`	net472	4.87μs	4.01ns	17.5ns	0.32	2.02 KB
#5968	`EnrichedLog`	net6.0	3.03μs	0.8ns	2.88ns	0.0305	2.2 KB
#5968	`EnrichedLog`	netcoreapp3.1	4.31μs	1.9ns	7.38ns	0.028	2.2 KB
#5968	`EnrichedLog`	net472	4.95μs	2.31ns	8.63ns	0.319	2.02 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`SendReceive`	net6.0	1.35μs	6.81ns	29.7ns	0.0159	0	1.14 KB
master	`SendReceive`	netcoreapp3.1	1.75μs	1.5ns	5.83ns	0.0159	0	1.14 KB
master	`SendReceive`	net472	2.05μs	0.992ns	3.84ns	0.183	0.00102	1.16 KB
#5968	`SendReceive`	net6.0	1.29μs	0.496ns	1.92ns	0.0161	0	1.14 KB
#5968	`SendReceive`	netcoreapp3.1	1.72μs	1.54ns	5.74ns	0.0156	0	1.14 KB
#5968	`SendReceive`	net472	2.13μs	0.823ns	3.08ns	0.183	0	1.16 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net6.0	2.9μs	0.864ns	3.35ns	0.022	1.6 KB
master	`EnrichedLog`	netcoreapp3.1	3.76μs	1.47ns	5.51ns	0.0226	1.65 KB
master	`EnrichedLog`	net472	4.48μs	2.72ns	10.5ns	0.323	2.04 KB
#5968	`EnrichedLog`	net6.0	2.87μs	0.944ns	3.66ns	0.0214	1.6 KB
#5968	`EnrichedLog`	netcoreapp3.1	3.83μs	2.03ns	7.88ns	0.021	1.65 KB
#5968	`EnrichedLog`	net472	4.38μs	2.89ns	11.2ns	0.324	2.04 KB

Benchmarks.Trace.SpanBenchmark - Slower ⚠️ Same allocations ✔️

Slower ⚠️ in #5968

Benchmark	diff/base	Base Median (ns)	Diff Median (ns)
Benchmarks.Trace.SpanBenchmark.StartFinishScope‑net6.0	1.256	473.60	595.06
Benchmarks.Trace.SpanBenchmark.StartFinishSpan‑net6.0	1.132	397.14	449.43
Benchmarks.Trace.SpanBenchmark.StartFinishSpan‑netcoreapp3.1	1.120	551.09	617.44
Benchmarks.Trace.SpanBenchmark.StartFinishSpan‑net472	1.116	587.07	655.28

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`StartFinishSpan`	net6.0	397ns	0.138ns	0.497ns	0.00815	576 B
master	`StartFinishSpan`	netcoreapp3.1	551ns	0.36ns	1.3ns	0.00772	576 B
master	`StartFinishSpan`	net472	587ns	0.495ns	1.92ns	0.0916	578 B
master	`StartFinishScope`	net6.0	473ns	0.369ns	1.43ns	0.00975	696 B
master	`StartFinishScope`	netcoreapp3.1	756ns	0.735ns	2.85ns	0.00944	696 B
master	`StartFinishScope`	net472	886ns	1.1ns	4.26ns	0.104	658 B
#5968	`StartFinishSpan`	net6.0	449ns	0.152ns	0.59ns	0.0081	576 B
#5968	`StartFinishSpan`	netcoreapp3.1	618ns	0.758ns	2.84ns	0.00776	576 B
#5968	`StartFinishSpan`	net472	656ns	1.13ns	4.37ns	0.0917	578 B
#5968	`StartFinishScope`	net6.0	595ns	0.103ns	0.384ns	0.00956	696 B
#5968	`StartFinishScope`	netcoreapp3.1	787ns	0.36ns	1.4ns	0.0095	696 B
#5968	`StartFinishScope`	net472	871ns	0.968ns	3.75ns	0.104	658 B

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`RunOnMethodBegin`	net6.0	658ns	0.173ns	0.649ns	0.00958	696 B
master	`RunOnMethodBegin`	netcoreapp3.1	958ns	0.556ns	2.15ns	0.00949	696 B
master	`RunOnMethodBegin`	net472	1.14μs	0.853ns	3.3ns	0.104	658 B
#5968	`RunOnMethodBegin`	net6.0	680ns	2.28ns	8.85ns	0.00978	696 B
#5968	`RunOnMethodBegin`	netcoreapp3.1	977ns	0.605ns	2.26ns	0.00953	696 B
#5968	`RunOnMethodBegin`	net472	1.1μs	0.759ns	2.94ns	0.104	658 B

gleocadie

LGTM

andrewlock · 2024-10-03T15:13:40Z

Throughput/Crank Report ⚡

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 (5968) (11.201M)   : 0, 11200519
    master (11.141M)   : 0, 11140702
    benchmarks/2.9.0 (11.081M)   : 0, 11080577

    section Automatic
    This PR (5968) (7.315M)   : 0, 7314888
    master (7.405M)   : 0, 7405201
    benchmarks/2.9.0 (7.732M)   : 0, 7732233

    section Trace stats
    master (7.560M)   : 0, 7559772

    section Manual
    master (11.167M)   : 0, 11167059

    section Manual + Automatic
    This PR (5968) (6.815M)   : 0, 6814790
    master (6.766M)   : 0, 6765500

    section DD_TRACE_ENABLED=0
    master (10.279M)   : 0, 10278615

gantt
    title Throughput Linux arm64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (5968) (9.669M)   : 0, 9668687
    master (9.721M)   : 0, 9721360
    benchmarks/2.9.0 (9.798M)   : 0, 9798067

    section Automatic
    This PR (5968) (6.608M)   : 0, 6607503
    master (6.623M)   : 0, 6623242

    section Trace stats
    master (6.765M)   : 0, 6764653

    section Manual
    master (9.661M)   : 0, 9660607

    section Manual + Automatic
    This PR (5968) (6.043M)   : 0, 6043397
    master (6.237M)   : 0, 6236560

    section DD_TRACE_ENABLED=0
    master (8.815M)   : 0, 8815374

gantt
    title Throughput Windows x64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (5968) (10.050M)   : 0, 10049713
    master (9.926M)   : 0, 9926042
    benchmarks/2.9.0 (10.067M)   : 0, 10067315

    section Automatic
    This PR (5968) (6.657M)   : 0, 6656563
    master (6.573M)   : 0, 6572681
    benchmarks/2.9.0 (7.552M)   : 0, 7552193

    section Trace stats
    master (7.374M)   : 0, 7373627

    section Manual
    master (10.104M)   : 0, 10103830

    section Manual + Automatic
    This PR (5968) (6.219M)   : 0, 6218654
    master (6.247M)   : 0, 6246669

    section DD_TRACE_ENABLED=0
    master (9.437M)   : 0, 9437343

This reverts commit 517046a.

Reverts #5968 It seems it causes a crash when .NET Framework and .NET Core are loaded in the same process.

) Reverts #5968 It seems it causes a crash when .NET Framework and .NET Core are loaded in the same process. Cherry pick of #6200 Co-authored-by: Kevin Gosse <kevin.gosse@datadoghq.com>

github-actions bot added the area:shared-components label Aug 30, 2024

kevingosse marked this pull request as ready for review August 30, 2024 12:20

kevingosse requested a review from a team as a code owner August 30, 2024 12:20

andrewlock approved these changes Aug 30, 2024

View reviewed changes

shared/src/Datadog.Trace.ClrProfiler.Native/dynamic_dispatcher.cpp Show resolved Hide resolved

tonyredondo approved these changes Aug 30, 2024

View reviewed changes

gleocadie approved these changes Sep 3, 2024

View reviewed changes

kevingosse force-pushed the kevin/better_loader branch from e1c10b8 to a1fdf2e Compare September 23, 2024 13:26

kevingosse force-pushed the kevin/better_loader branch from a1fdf2e to 3a9afd9 Compare October 2, 2024 15:13

kevingosse added 4 commits October 3, 2024 12:33

Load the tracer/profiler only during CorProfilerInfo::Initialize

1bab6c5

Fix tests

684673d

Update dynamic_dispatcher.cpp

075d517

Fix error checking

07e6afd

kevingosse force-pushed the kevin/better_loader branch from 3a9afd9 to 07e6afd Compare October 3, 2024 12:00

kevingosse merged commit 517046a into master Oct 4, 2024
82 checks passed

kevingosse deleted the kevin/better_loader branch October 4, 2024 13:05

github-actions bot added this to the vNext-v3 milestone Oct 4, 2024

kevingosse added a commit that referenced this pull request Oct 25, 2024

Revert "Load the tracer/profiler after guardrails checks (#5968)"

7264f85

This reverts commit 517046a.

kevingosse mentioned this pull request Oct 25, 2024

Revert "Load the tracer/profiler after guardrails checks" #6200

Merged

kevingosse added a commit that referenced this pull request Oct 25, 2024

Revert "Load the tracer/profiler after guardrails checks" (#6200)

4d9df8b

Reverts #5968 It seems it causes a crash when .NET Framework and .NET Core are loaded in the same process.

andrewlock pushed a commit that referenced this pull request Oct 25, 2024

Revert "Load the tracer/profiler after guardrails checks" (#6200)

f6a1f76

Reverts #5968 It seems it causes a crash when .NET Framework and .NET Core are loaded in the same process.

andrewlock mentioned this pull request Oct 25, 2024

Revert "Load the tracer/profiler after guardrails checks" (#6200) #6205

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Load the tracer/profiler after guardrails checks #5968

Load the tracer/profiler after guardrails checks #5968

kevingosse commented Aug 30, 2024 •

edited

Loading

andrewlock left a comment

datadog-ddstaging bot commented Aug 30, 2024 •

edited

Loading

andrewlock commented Sep 2, 2024 •

edited

Loading

andrewlock commented Sep 2, 2024 •

edited

Loading

Raw results

Raw results

Raw results

More allocations ⚠️ in #5968

Raw results

Raw results

Faster 🎉 in #5968

Raw results

Slower ⚠️ in #5968

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Slower ⚠️ in #5968

Raw results

Raw results

gleocadie left a comment

andrewlock commented Oct 3, 2024

Load the tracer/profiler after guardrails checks #5968

Load the tracer/profiler after guardrails checks #5968

Conversation

kevingosse commented Aug 30, 2024 • edited Loading

Summary of changes

Reason for change

Implementation details

Test coverage

Other details

andrewlock left a comment

Choose a reason for hiding this comment

datadog-ddstaging bot commented Aug 30, 2024 • edited Loading

Datadog Report

New Flaky Tests (1)

andrewlock commented Sep 2, 2024 • edited Loading

Execution-Time Benchmarks Report ⏱️

andrewlock commented Sep 2, 2024 • edited Loading

Benchmarks Report for tracer 🐌

Benchmark details

Raw results

Raw results

Raw results

More allocations ⚠️ in #5968

Raw results

Raw results

Faster 🎉 in #5968

Raw results

Slower ⚠️ in #5968

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Slower ⚠️ in #5968

Raw results

Raw results

gleocadie left a comment

Choose a reason for hiding this comment

andrewlock commented Oct 3, 2024

Throughput/Crank Report ⚡

kevingosse commented Aug 30, 2024 •

edited

Loading

datadog-ddstaging bot commented Aug 30, 2024 •

edited

Loading

andrewlock commented Sep 2, 2024 •

edited

Loading

andrewlock commented Sep 2, 2024 •

edited

Loading