Fix assumptions about what a top_level span is

[bwoebi]

Otherwise span stats will be broken for nested services.

Also properly handling version propagation, its removal according to UST was too aggressive until now.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 531586746b

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

[chatgpt-codex-connector]

Use resolved version when keying concentrator writes

ddtrace_feed_span_to_concentrator now ignores pre->version and reads only span->root->property_version when it is already a string, which drops versions that were resolved via deprecated meta["version"] (still supported in ddtrace_precompute_span) or via non-string values converted during precompute. In those cases spans are emitted with a version tag but stats are keyed with an empty version, causing version-level stats to be mis-grouped or split from trace data.

Useful? React with 👍 / 👎.

[datadog-prod-us1-3]

 

✨ Fix all issues with BitsAI

⚠️ Warnings

🚦 47 Pipeline jobs failed

DataDog/apm-reliability/dd-trace-php | ASAN test_c: [7.4, arm64]     

🔧 Fix in code (Fix with Cursor).

Command terminated with exit code 2 during ASAN test execution.

🧪 1 Test failed

tmp/build_extension/tests/ext/ust.phpt (Foo) from PHP.tmp.build_extension.tests.ext   (Fix with Cursor)

030+       ["version"]=>
031+       string(5) "5.2.0"

DataDog/apm-reliability/dd-trace-php | ASAN test_c: [8.5, arm64]     

🔧 Fix in code (Fix with Cursor).

Test command failed with exit code 2 due to make error in the test_c target.

🧪 1 Test failed

tmp/build_extension/tests/ext/ust.phpt (Foo) from php.tmp.build_extension.tests.ext   (Fix with Cursor)

--
           string(8) "env_test"
           ["runtime-id"]=>
           string(36) "%s"
030+       ["version"]=>
031+       string(5) "5.2.0"
         }
         ["metrics"]=>
         array(%d) {
--

DataDog/apm-reliability/dd-trace-php | System Tests: [parametric]     

🔧 Fix in code (Fix with Cursor).

1 failed test due to assertion error: assert 'version' not in span2['meta']

🧪 1 Test failed

tests.parametric.test_config_consistency.Test_Config_UnifiedServiceTagging.test_specific_version[library_env0, parametric-php] from system_tests_suite   (Fix with Cursor)

AssertionError: assert 'version' not in {'_dd.p.dm': '-0', '_dd.p.tid': '6a14715c00000000', '_dd.tags.process': 'entrypoint.basedir:binaries,entrypoint.name:s...t.type:script,entrypoint.workdir:binaries,runtime.sapi:cli', 'runtime-id': '72b10d8d-601e-4ee5-91ab-6602753638a6', ...}

self = <tests.parametric.test_config_consistency.Test_Config_UnifiedServiceTagging object at 0x7fd1ae5894c0>
test_agent = <utils.docker_fixtures._test_agent.TestAgentAPI object at 0x7fd185bdb2c0>
test_library = <utils.docker_fixtures._test_clients._test_client_parametric.ParametricTestClientApi object at 0x7fd1865abb30>

    @parametrize("library_env", [{"DD_SERVICE": "version_test", "DD_VERSION": "5.2.0"}])
    def test_specific_version(self, test_agent: TestAgentAPI, test_library: APMLibrary):
        with test_library:
            with test_library.dd_start_span(name="s1") as s1:
...

View all 47 failed jobs.

ℹ️ Info

No other issues found (see more)

❄️ No new flaky tests detected

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 60.70% (-0.05%)

Useful? React with 👍 / 👎

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 5315867 | Docs | Datadog PR Page | Give us feedback!}

[pr-commenter]

Benchmarks [ tracer ]

Benchmark execution time: 2026-05-25 16:40:49

Comparing candidate commit 5315867 in PR branch bob/fix-span-stats with baseline commit 0fb4c56 in branch master.

Found 0 performance improvements and 14 performance regressions! Performance is the same for 178 metrics, 2 unstable metrics.

scenario:PDOBench/benchPDOOverhead

• 🟥 execution_time [+11.019µs; +13.417µs] or [+4.477%; +5.452%]

scenario:PDOBench/benchPDOOverhead-opcache

• 🟥 execution_time [+12.974µs; +14.846µs] or [+5.316%; +6.083%]

scenario:PDOBench/benchPDOOverheadWithDBM

• 🟥 execution_time [+10.846µs; +13.490µs] or [+4.412%; +5.488%]

scenario:PDOBench/benchPDOOverheadWithDBM-opcache

• 🟥 execution_time [+12.006µs; +14.084µs] or [+4.915%; +5.766%]

scenario:PHPRedisBench/benchRedisOverhead

• 🟥 execution_time [+51.400µs; +69.178µs] or [+5.375%; +7.234%]

scenario:PHPRedisBench/benchRedisOverhead-opcache

• 🟥 execution_time [+56.329µs; +71.108µs] or [+5.632%; +7.110%]

scenario:SpanBench/benchDatadogAPI

• 🟥 execution_time [+4.812µs; +6.296µs] or [+7.480%; +9.788%]

scenario:SpanBench/benchDatadogAPI-opcache

• 🟥 execution_time [+4.909µs; +5.752µs] or [+7.595%; +8.900%]

scenario:SpanBench/benchOpenTelemetryInteroperability

• 🟥 execution_time [+4.724µs; +7.261µs] or [+2.505%; +3.850%]

scenario:SpanBench/benchOpenTelemetryInteroperability-opcache

• 🟥 execution_time [+81.080µs; +85.228µs] or [+28.622%; +30.086%]

scenario:TraceFlushBench/benchFlushTrace

• 🟥 execution_time [+53.371µs; +91.629µs] or [+4.570%; +7.846%]

scenario:TraceFlushBench/benchFlushTrace-opcache

• 🟥 execution_time [+71.057µs; +93.143µs] or [+6.066%; +7.952%]

scenario:TraceSerializationBench/benchSerializeTrace

• 🟥 execution_time [+94.875µs; +111.725µs] or [+22.730%; +26.767%]

scenario:TraceSerializationBench/benchSerializeTrace-opcache

• 🟥 execution_time [+96.650µs; +110.750µs] or [+27.925%; +32.000%]