[Profiler] Fix crash at shutdown with the `timer_create`-based CPU profiler #6268

gleocadie · 2024-11-10T18:05:34Z

Summary of changes

Make sure the profiler does not crash at shutdown.

Reason for change

For the timer_create-based CPU profiler, each managed thread is associated to timer so that every 9ms, a callback is called and we collect the thread snapshot (metadata + callstack). To store in a signal-safe manner the thread snapshot we use a synchronized_pool_resource(https://en.cppreference.com/w/cpp/memory/synchronized_pool_resource) + mmap as an upstream (when the pool needs more memory, it calls into mmap and not malloc). As an optimization, this resource uses a thread local variable so each native thread has its own pools (reduce lock contention).

When an application is shutting down, the CLR follows a sequence of steps to stop and clean the resources it used. One of those steps is to notify the ICorprofiler profilers (through the callbacks). When a managed thread is terminated, it cleans up the associated resource, call the ThreadDestroyed callback... When the CLR is done with the thread clean up (the routine passed to the native thread ends), the libc takes on and cleans up the native resources (clean up thread local storage...).

But sometimes, those ThreadDestroyed is not called and it happens when the Execution Engine has started to shutdown https://github.com/dotnet/runtime/blob/main/src/coreclr/vm/threads.cpp#L2777
This leads up to a crash situation where:

The timer associated is still here
Libc is destroying thread local storage for this thread
The timer kicks in, starts collecting the thread snapshot and use the pools associated to the thread (which are being destroyed) to store it.

Implementation details

Use the pthread_create wrapping to wrap the thread entrypoint.
Add a callback into CorProfilerCallback to do extra cleanup.

Test coverage

Current tests should not crash at shutdown

Other details

datadog-ddstaging · 2024-11-10T18:34:41Z

Datadog Report

Branch report: gleocadie/stop-profiler-in-case-of-crash
Commit report: e0b4bb8
Test service: dd-trace-dotnet

❌ 1 Failed (1 Known Flaky), 370728 Passed, 2397 Skipped, 16h 45m 54.1s Total Time
❄️ 1 New Flaky

❌ Failed Tests (1)

ShouldGetLockContentionSamplesViaEtw - Datadog.Profiler.IntegrationTests.Contention.ContentionProfilerTest - ❄️ Known flaky - Details
Expand for error
```
 
 Actual:   Falseilure
```

New Flaky Tests (1)

CanCreateDogStatsD_UDP_FromTraceAgentSettings - Datadog.Trace.Tests.DogStatsDTests - Last Failure
Expand for error
```
 Address in use
```

andrewlock · 2024-11-10T18:35:33Z

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 (6268) - mean (70ms)  : 68, 72
     .   : milestone, 70,
    master - mean (70ms)  : 68, 71
     .   : milestone, 70,

    section CallTarget+Inlining+NGEN
    This PR (6268) - mean (1,118ms)  : 1101, 1136
     .   : milestone, 1118,
    master - mean (1,116ms)  : 1099, 1133
     .   : milestone, 1116,

gantt
    title Execution time (ms) FakeDbCommand (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (6268) - mean (108ms)  : 106, 110
     .   : milestone, 108,
    master - mean (108ms)  : 106, 110
     .   : milestone, 108,

    section CallTarget+Inlining+NGEN
    This PR (6268) - mean (774ms)  : 761, 788
     .   : milestone, 774,
    master - mean (775ms)  : 762, 787
     .   : milestone, 775,

gantt
    title Execution time (ms) FakeDbCommand (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (6268) - mean (92ms)  : 90, 94
     .   : milestone, 92,
    master - mean (91ms)  : 90, 93
     .   : milestone, 91,

    section CallTarget+Inlining+NGEN
    This PR (6268) - mean (727ms)  : 712, 743
     .   : milestone, 727,
    master - mean (731ms)  : 716, 746
     .   : milestone, 731,

gantt
    title Execution time (ms) HttpMessageHandler (.NET Framework 4.6.2) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (6268) - mean (189ms)  : 186, 193
     .   : milestone, 189,
    master - mean (191ms)  : 185, 196
     .   : milestone, 191,

    section CallTarget+Inlining+NGEN
    This PR (6268) - mean (1,226ms)  : 1200, 1252
     .   : milestone, 1226,
    master - mean (1,229ms)  : 1208, 1250
     .   : milestone, 1229,

gantt
    title Execution time (ms) HttpMessageHandler (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (6268) - mean (276ms)  : 272, 279
     .   : milestone, 276,
    master - mean (275ms)  : 270, 280
     .   : milestone, 275,

    section CallTarget+Inlining+NGEN
    This PR (6268) - mean (953ms)  : 932, 973
     .   : milestone, 953,
    master - mean (949ms)  : 934, 964
     .   : milestone, 949,

gantt
    title Execution time (ms) HttpMessageHandler (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (6268) - mean (264ms)  : 260, 268
     .   : milestone, 264,
    master - mean (264ms)  : 260, 268
     .   : milestone, 264,

    section CallTarget+Inlining+NGEN
    This PR (6268) - mean (927ms)  : 912, 943
     .   : milestone, 927,
    master - mean (935ms)  : 913, 957
     .   : milestone, 935,

andrewlock · 2024-11-10T19:37:20Z

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 (6268) (11.158M)   : 0, 11158423
    master (11.152M)   : 0, 11152138
    benchmarks/2.9.0 (11.033M)   : 0, 11032866

    section Automatic
    This PR (6268) (7.280M)   : 0, 7280331
    master (7.262M)   : 0, 7261734
    benchmarks/2.9.0 (7.786M)   : 0, 7785853

    section Trace stats
    master (7.549M)   : 0, 7548778

    section Manual
    master (11.173M)   : 0, 11173023

    section Manual + Automatic
    This PR (6268) (6.781M)   : 0, 6781201
    master (6.794M)   : 0, 6793544

    section DD_TRACE_ENABLED=0
    master (10.214M)   : 0, 10214196

gantt
    title Throughput Linux arm64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (6268) (9.751M)   : 0, 9751363
    master (9.531M)   : 0, 9530515
    benchmarks/2.9.0 (9.495M)   : 0, 9494821

    section Automatic
    This PR (6268) (6.497M)   : 0, 6496948
    master (6.486M)   : 0, 6486418

    section Trace stats
    master (6.867M)   : 0, 6867398

    section Manual
    master (9.426M)   : 0, 9425938

    section Manual + Automatic
    This PR (6268) (6.009M)   : 0, 6008606
    master (6.122M)   : 0, 6121877

    section DD_TRACE_ENABLED=0
    master (8.820M)   : 0, 8820217

gantt
    title Throughput Windows x64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (6268) (9.761M)   : 0, 9761440
    master (9.876M)   : 0, 9875828
    benchmarks/2.9.0 (10.020M)   : 0, 10019592

    section Automatic
    This PR (6268) (6.314M)   : 0, 6314238
    master (6.174M)   : 0, 6174285
    benchmarks/2.9.0 (7.255M)   : 0, 7255257

    section Trace stats
    master (6.971M)   : 0, 6971003

    section Manual
    master (9.416M)   : 0, 9415959

    section Manual + Automatic
    This PR (6268) (5.955M)   : 0, 5955291
    master (5.884M)   : 0, 5883842

    section DD_TRACE_ENABLED=0
    master (8.894M)   : 0, 8893965

andrewlock · 2024-11-10T19:37:43Z

Benchmarks Report for tracer 🐌

Benchmarks for #6268 compared to master:

2 benchmarks are faster, with geometric mean 1.136
1 benchmarks are slower, with geometric mean 1.126
All benchmarks have the same 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	7.99μs	44.2ns	283ns	0.0152	0.00759	0	5.6 KB
master	`StartStopWithChild`	netcoreapp3.1	10.3μs	55.4ns	304ns	0.0153	0.00509	0	5.8 KB
master	`StartStopWithChild`	net472	16.4μs	39.9ns	149ns	1.03	0.286	0.0816	6.21 KB
#6268	`StartStopWithChild`	net6.0	8.12μs	45.1ns	282ns	0.0196	0.0117	0.00392	5.61 KB
#6268	`StartStopWithChild`	netcoreapp3.1	10.1μs	56.2ns	355ns	0.0192	0.0096	0	5.8 KB
#6268	`StartStopWithChild`	net472	16.3μs	43.9ns	164ns	1.06	0.322	0.0886	6.21 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`WriteAndFlushEnrichedTraces`	net6.0	478μs	274ns	1.06μs	0	2.7 KB
master	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	651μs	401ns	1.45μs	0	2.7 KB
master	`WriteAndFlushEnrichedTraces`	net472	841μs	833ns	3.23μs	0.422	3.3 KB
#6268	`WriteAndFlushEnrichedTraces`	net6.0	470μs	215ns	831ns	0	2.7 KB
#6268	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	652μs	234ns	844ns	0	2.7 KB
#6268	`WriteAndFlushEnrichedTraces`	net472	830μs	186ns	646ns	0.414	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	202μs	1.16μs	9.03μs	0.197	18.73 KB
master	`SendRequest`	netcoreapp3.1	222μs	1.26μs	9.17μs	0.214	20.89 KB
master	`SendRequest`	net472	0.0211ns	0.00519ns	0.0201ns	0	0 b
#6268	`SendRequest`	net6.0	196μs	1.12μs	8.36μs	0.204	18.73 KB
#6268	`SendRequest`	netcoreapp3.1	209μs	1.02μs	4.34μs	0.227	20.89 KB
#6268	`SendRequest`	net472	0.000563ns	0.000381ns	0.00147ns	0	0 b

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Gen 2	Allocated
master	`WriteAndFlushEnrichedTraces`	net6.0	565μs	2.42μs	9.37μs	0.579	0	0	41.57 KB
master	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	680μs	3.6μs	17.7μs	0.336	0	0	41.87 KB
master	`WriteAndFlushEnrichedTraces`	net472	856μs	3.36μs	13μs	8.25	2.6	0.434	53.32 KB
#6268	`WriteAndFlushEnrichedTraces`	net6.0	566μs	2.98μs	14.6μs	0.563	0	0	41.72 KB
#6268	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	689μs	3.72μs	19.7μs	0.324	0	0	41.91 KB
#6268	`WriteAndFlushEnrichedTraces`	net472	861μs	4.29μs	18.7μs	8.08	2.55	0.425	53.32 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`ExecuteNonQuery`	net6.0	1.26μs	0.873ns	3.27ns	0.0142	0	1.02 KB
master	`ExecuteNonQuery`	netcoreapp3.1	1.71μs	3.27ns	12.2ns	0.0137	0	1.02 KB
master	`ExecuteNonQuery`	net472	2.09μs	1.39ns	5.22ns	0.156	0.00105	987 B
#6268	`ExecuteNonQuery`	net6.0	1.26μs	0.663ns	2.57ns	0.0139	0	1.02 KB
#6268	`ExecuteNonQuery`	netcoreapp3.1	1.82μs	0.503ns	1.74ns	0.0135	0	1.02 KB
#6268	`ExecuteNonQuery`	net472	2.11μs	1.36ns	5.28ns	0.157	0.00105	987 B

Benchmarks.Trace.ElasticsearchBenchmark - Slower ⚠️ Same allocations ✔️

Slower ⚠️ in #6268

Benchmark	diff/base	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.ElasticsearchBenchmark.CallElasticsearch‑net6.0	1.126	1,172.70	1,320.35

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`CallElasticsearch`	net6.0	1.17μs	0.384ns	1.38ns	0.0135	976 B
master	`CallElasticsearch`	netcoreapp3.1	1.54μs	2.2ns	8.52ns	0.0133	976 B
master	`CallElasticsearch`	net472	2.69μs	2.28ns	8.85ns	0.158	995 B
master	`CallElasticsearchAsync`	net6.0	1.38μs	0.616ns	2.39ns	0.0131	952 B
master	`CallElasticsearchAsync`	netcoreapp3.1	1.71μs	0.599ns	2.16ns	0.0137	1.02 KB
master	`CallElasticsearchAsync`	net472	2.68μs	0.98ns	3.67ns	0.166	1.05 KB
#6268	`CallElasticsearch`	net6.0	1.32μs	0.378ns	1.36ns	0.0139	976 B
#6268	`CallElasticsearch`	netcoreapp3.1	1.67μs	1.14ns	4.4ns	0.0134	976 B
#6268	`CallElasticsearch`	net472	2.61μs	2.98ns	10.7ns	0.157	995 B
#6268	`CallElasticsearchAsync`	net6.0	1.34μs	0.616ns	2.39ns	0.0134	952 B
#6268	`CallElasticsearchAsync`	netcoreapp3.1	1.6μs	1.27ns	4.94ns	0.0136	1.02 KB
#6268	`CallElasticsearchAsync`	net472	2.61μs	2.09ns	8.08ns	0.166	1.05 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`ExecuteAsync`	net6.0	1.37μs	0.819ns	3.06ns	0.0131	952 B
master	`ExecuteAsync`	netcoreapp3.1	1.71μs	0.913ns	3.42ns	0.0128	952 B
master	`ExecuteAsync`	net472	1.85μs	0.496ns	1.92ns	0.145	915 B
#6268	`ExecuteAsync`	net6.0	1.26μs	0.581ns	2.25ns	0.0133	952 B
#6268	`ExecuteAsync`	netcoreapp3.1	1.58μs	1.16ns	4.5ns	0.0126	952 B
#6268	`ExecuteAsync`	net472	1.81μs	0.796ns	2.98ns	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	4.4μs	2.26ns	8.15ns	0.0331	2.31 KB
master	`SendAsync`	netcoreapp3.1	5.19μs	6.13ns	23.7ns	0.0382	2.85 KB
master	`SendAsync`	net472	7.5μs	1.98ns	7.69ns	0.494	3.12 KB
#6268	`SendAsync`	net6.0	4.42μs	2.55ns	9.55ns	0.0331	2.31 KB
#6268	`SendAsync`	netcoreapp3.1	5.34μs	1.75ns	6.77ns	0.0374	2.85 KB
#6268	`SendAsync`	net472	7.35μs	2.24ns	8.66ns	0.493	3.12 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net6.0	1.52μs	1.19ns	4.47ns	0.0234	1.64 KB
master	`EnrichedLog`	netcoreapp3.1	2.16μs	0.869ns	3.13ns	0.0225	1.64 KB
master	`EnrichedLog`	net472	2.59μs	3ns	11.6ns	0.249	1.57 KB
#6268	`EnrichedLog`	net6.0	1.53μs	0.433ns	1.62ns	0.023	1.64 KB
#6268	`EnrichedLog`	netcoreapp3.1	2.18μs	1.3ns	4.85ns	0.0218	1.64 KB
#6268	`EnrichedLog`	net472	2.7μs	2.26ns	8.76ns	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	118μs	152ns	590ns	0	0	4.28 KB
master	`EnrichedLog`	netcoreapp3.1	123μs	176ns	681ns	0	0	4.28 KB
master	`EnrichedLog`	net472	152μs	151ns	585ns	0.675	0.225	4.46 KB
#6268	`EnrichedLog`	net6.0	118μs	211ns	816ns	0.0591	0	4.28 KB
#6268	`EnrichedLog`	netcoreapp3.1	123μs	295ns	1.14μs	0.0608	0	4.28 KB
#6268	`EnrichedLog`	net472	154μs	147ns	569ns	0.694	0.231	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	3.05μs	1.91ns	7.39ns	0.0305	2.2 KB
master	`EnrichedLog`	netcoreapp3.1	4.08μs	0.894ns	3.46ns	0.0285	2.2 KB
master	`EnrichedLog`	net472	4.84μs	1.45ns	5.61ns	0.32	2.02 KB
#6268	`EnrichedLog`	net6.0	3.06μs	1.4ns	5.24ns	0.0306	2.2 KB
#6268	`EnrichedLog`	netcoreapp3.1	4.23μs	1.11ns	4.29ns	0.0296	2.2 KB
#6268	`EnrichedLog`	net472	4.87μs	0.965ns	3.74ns	0.32	2.02 KB

Benchmarks.Trace.RedisBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #6268

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.RedisBenchmark.SendReceive‑net472	1.137	2,302.87	2,025.78

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendReceive`	net6.0	1.5μs	1.01ns	3.92ns	0.0158	1.14 KB
master	`SendReceive`	netcoreapp3.1	1.79μs	0.92ns	3.44ns	0.0151	1.14 KB
master	`SendReceive`	net472	2.3μs	1.64ns	6.34ns	0.183	1.16 KB
#6268	`SendReceive`	net6.0	1.41μs	1.15ns	4.44ns	0.0161	1.14 KB
#6268	`SendReceive`	netcoreapp3.1	1.75μs	1.43ns	5.54ns	0.0157	1.14 KB
#6268	`SendReceive`	net472	2.03μs	0.893ns	3.22ns	0.183	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	2.69μs	0.725ns	2.81ns	0.0215	1.6 KB
master	`EnrichedLog`	netcoreapp3.1	3.85μs	1.5ns	5.79ns	0.0213	1.65 KB
master	`EnrichedLog`	net472	4.45μs	3.04ns	11.8ns	0.322	2.04 KB
#6268	`EnrichedLog`	net6.0	2.71μs	0.666ns	2.49ns	0.0231	1.6 KB
#6268	`EnrichedLog`	netcoreapp3.1	4.01μs	2.09ns	8.1ns	0.022	1.65 KB
#6268	`EnrichedLog`	net472	4.75μs	2.99ns	11.6ns	0.324	2.04 KB

Benchmarks.Trace.SpanBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #6268

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.SpanBenchmark.StartFinishScope‑net6.0	1.135	558.01	491.85

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`StartFinishSpan`	net6.0	398ns	0.469ns	1.76ns	0.00798	576 B
master	`StartFinishSpan`	netcoreapp3.1	634ns	0.471ns	1.83ns	0.00767	576 B
master	`StartFinishSpan`	net472	676ns	0.662ns	2.56ns	0.0917	578 B
master	`StartFinishScope`	net6.0	558ns	0.362ns	1.4ns	0.00972	696 B
master	`StartFinishScope`	netcoreapp3.1	745ns	0.72ns	2.79ns	0.00963	696 B
master	`StartFinishScope`	net472	904ns	0.487ns	1.89ns	0.104	658 B
#6268	`StartFinishSpan`	net6.0	411ns	0.189ns	0.733ns	0.00804	576 B
#6268	`StartFinishSpan`	netcoreapp3.1	570ns	0.667ns	2.58ns	0.00775	576 B
#6268	`StartFinishSpan`	net472	693ns	0.613ns	2.37ns	0.0917	578 B
#6268	`StartFinishScope`	net6.0	492ns	0.388ns	1.5ns	0.00964	696 B
#6268	`StartFinishScope`	netcoreapp3.1	760ns	0.671ns	2.6ns	0.00941	696 B
#6268	`StartFinishScope`	net472	910ns	1.83ns	7.09ns	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	666ns	1.55ns	5.99ns	0.00981	696 B
master	`RunOnMethodBegin`	netcoreapp3.1	893ns	2.04ns	7.62ns	0.00932	696 B
master	`RunOnMethodBegin`	net472	1.15μs	1.27ns	4.93ns	0.104	658 B
#6268	`RunOnMethodBegin`	net6.0	611ns	0.497ns	1.93ns	0.00981	696 B
#6268	`RunOnMethodBegin`	netcoreapp3.1	928ns	0.859ns	3.33ns	0.00931	696 B
#6268	`RunOnMethodBegin`	net472	1.16μs	1.09ns	4.2ns	0.104	658 B

chrisnas

LGTM

github-actions bot added the area:profiler Issues related to the continous-profiler label Nov 10, 2024

gleocadie force-pushed the gleocadie/stop-profiler-in-case-of-crash branch from 352e5e7 to e011b72 Compare November 10, 2024 18:38

gleocadie force-pushed the gleocadie/stop-profiler-in-case-of-crash branch 2 times, most recently from b8afe0d to 449cfb1 Compare November 10, 2024 22:06

Fix crash at shutdown

30e6d39

gleocadie force-pushed the gleocadie/stop-profiler-in-case-of-crash branch 3 times, most recently from 15b90e3 to c0a1e84 Compare November 12, 2024 13:00

Add a callback into the profiler

e0b4bb8

gleocadie force-pushed the gleocadie/stop-profiler-in-case-of-crash branch from c0a1e84 to e0b4bb8 Compare November 12, 2024 15:20

gleocadie changed the title ~~[Profiler] Fix crash at shutdown~~ [Profiler] Fix crash at shutdown with the timer_create-based CPU profiler Nov 12, 2024

gleocadie marked this pull request as ready for review November 12, 2024 16:23

gleocadie requested a review from a team as a code owner November 12, 2024 16:23

chrisnas approved these changes Nov 13, 2024

View reviewed changes

gleocadie merged commit a04e38b into master Nov 13, 2024
82 of 85 checks passed

gleocadie deleted the gleocadie/stop-profiler-in-case-of-crash branch November 13, 2024 16:10

github-actions bot added this to the vNext-v3 milestone Nov 13, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Profiler] Fix crash at shutdown with the `timer_create`-based CPU profiler #6268

[Profiler] Fix crash at shutdown with the `timer_create`-based CPU profiler #6268

gleocadie commented Nov 10, 2024 •

edited

Loading

datadog-ddstaging bot commented Nov 10, 2024 •

edited

Loading

andrewlock commented Nov 10, 2024 •

edited

Loading

andrewlock commented Nov 10, 2024 •

edited

Loading

andrewlock commented Nov 10, 2024 •

edited

Loading

Raw results

Raw results

Raw results

Raw results

Raw results

Slower ⚠️ in #6268

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Faster 🎉 in #6268

Raw results

Raw results

Faster 🎉 in #6268

Raw results

Raw results

chrisnas left a comment

[Profiler] Fix crash at shutdown with the timer_create-based CPU profiler #6268

[Profiler] Fix crash at shutdown with the timer_create-based CPU profiler #6268

Conversation

gleocadie commented Nov 10, 2024 • edited Loading

Summary of changes

Reason for change

Implementation details

Test coverage

Other details

datadog-ddstaging bot commented Nov 10, 2024 • edited Loading

Datadog Report

❌ Failed Tests (1)

New Flaky Tests (1)

andrewlock commented Nov 10, 2024 • edited Loading

Execution-Time Benchmarks Report ⏱️

andrewlock commented Nov 10, 2024 • edited Loading

Throughput/Crank Report ⚡

andrewlock commented Nov 10, 2024 • edited Loading

Benchmarks Report for tracer 🐌

Benchmark details

Raw results

Raw results

Raw results

Raw results

Raw results

Slower ⚠️ in #6268

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Faster 🎉 in #6268

Raw results

Raw results

Faster 🎉 in #6268

Raw results

Raw results

chrisnas left a comment

Choose a reason for hiding this comment

[Profiler] Fix crash at shutdown with the `timer_create`-based CPU profiler #6268

[Profiler] Fix crash at shutdown with the `timer_create`-based CPU profiler #6268

gleocadie commented Nov 10, 2024 •

edited

Loading

datadog-ddstaging bot commented Nov 10, 2024 •

edited

Loading

andrewlock commented Nov 10, 2024 •

edited

Loading

andrewlock commented Nov 10, 2024 •

edited

Loading

andrewlock commented Nov 10, 2024 •

edited

Loading