feat(autoinstrumentation): compile SSI targets to policies and wire the mutator to the policy engine

[iamluc]

What does this PR do?

Evolves Kubernetes SSI targets matching from k8s label selectors to the
shared Go policy engine (github.com/DataDog/dd-policy-engine/go), so the
cluster-agent shares one rule semantics with the host injector.

Three commits, in order:

1. test — a behavioral characterization (golden master) of the target
   matcher, exercised only through the public TargetMutator API
   (NewTargetMutator + getMatchingTarget). Added before the refactor so it
   runs unchanged against the current label-selector implementation.
2. compiler — adds the dd-policy-engine/go dependency and lowers the
   static targets config into policies (policiesFromTargets).
3. wiring — the TargetMutator matches pods via the policy engine
   (policyMatcher) instead of label selectors; targets and policies stay
   aligned by index, preserving first-match-wins.

Scope is intentionally minimal: no remote-config path is wired here (that
follows in a separate PR). NewTargetMutator keeps its signature, so callers
and existing tests are unchanged.

Matching coverage (characterization)

first-match-wins precedence · pod matchLabels (ANDed) · pod matchExpressions
In/NotIn/Exists/DoesNotExist (incl. "absent key matches NotIn") · namespace
matchNames/matchLabels/matchExpressions · combined namespace+pod (ANDed) ·
empty-target-matches-all · disabled-namespace short-circuit.

The suite passes against both the legacy label-selector implementation
(verified on main) and the policy-engine implementation, proving behavior is
preserved.

Intentional behavior change

target_matching_change_test.go pins one deliberate change: when a pod's
namespace is absent from the workloadmeta store and a target reads namespace
labels, the matcher now declines injection consistently (the legacy matcher
was order-dependent and could still match an earlier pod-only target). This test
is red on the legacy impl and green on the new one, documenting the change.

Describe how you validated your changes

go build -tags kubeapiserver ./pkg/clusteragent/admission/mutate/autoinstrumentation/...
go test -tags "test kubeapiserver" ./pkg/clusteragent/admission/mutate/autoinstrumentation/

Characterization tests were additionally run against main (legacy impl) via a
git worktree to confirm equivalence.

Additional Notes

• The dependency is pinned to a pseudo-version of the dd-policy-engine PR
  branch (feat(go): native Go policy evaluator for Kubernetes SSI targeting dd-policy-engine#71, draft). Repin to the released version
  once that PR merges — this PR is stacked on it and stays in draft until then.
• golang.org/x/sys is bumped 0.45.0 -> 0.46.0 as required by the
  dependency's go.mod (MVS).

[datadog-datadog-prod-us1]

✨ Fix all issues with BitsAI

⚠️ Warnings

🚦 2 Pipeline jobs failed

Label analysis | release-note-check     

Label analysis | skip-qa-check     

ℹ️ Info

🎯 Code Coverage (details)
• Patch Coverage: 92.66%
• Overall Coverage: 50.93% (+0.01%)

Useful? React with 👍 / 👎

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 5acd305 | Docs | Datadog PR Page | Give us feedback!}

[dd-octo-sts]

Go Package Import Differences

Baseline: aced8ce
Comparison: 5acd305

binary	os	arch	change
cluster-agent	linux	amd64	+1, -0 +github.com/DataDog/dd-policy-engine/go/policies
cluster-agent	linux	arm64	+1, -0 +github.com/DataDog/dd-policy-engine/go/policies

[dd-octo-sts]

Files inventory check summary

File checks results against ancestor aced8ce7:

Results for datadog-agent_7.82.0~devel.git.180.5acd305.pipeline.120124111-1_amd64.deb:

No change detected

[dd-octo-sts]

Static quality checks

✅ Please find below the results from static quality gates
Comparison made with ancestor aced8ce
📊 Static Quality Gates Dashboard
🔗 SQG Job

Successful checks

Info

	Quality gate	Change	Size (prev → curr → max)
✅	agent_deb_amd64	+10.03 KiB (0.00% increase, -0.11% of buffer)	749.074 → 749.084 → 758.200
✅	agent_deb_amd64_fips	-9.72 KiB (0.00% reduction, +0.20% of buffer)	705.050 → 705.040 → 709.840
✅	agent_rpm_amd64	+10.03 KiB (0.00% increase, -0.11% of buffer)	749.058 → 749.067 → 758.170
✅	agent_rpm_amd64_fips	-9.72 KiB (0.00% reduction, +0.20% of buffer)	705.033 → 705.024 → 709.840
✅	agent_suse_amd64	+10.03 KiB (0.00% increase, -0.11% of buffer)	749.058 → 749.067 → 758.170
✅	agent_suse_amd64_fips	-9.72 KiB (0.00% reduction, +0.20% of buffer)	705.033 → 705.024 → 709.840
✅	docker_agent_amd64	+10.03 KiB (0.00% increase, -0.20% of buffer)	808.812 → 808.822 → 813.790
✅	docker_agent_jmx_amd64	+10.03 KiB (0.00% increase, -0.20% of buffer)	999.709 → 999.719 → 1004.550
✅	docker_cluster_agent_amd64	+24.09 KiB (0.01% increase, -1.06% of buffer)	208.241 → 208.264 → 210.470
✅	docker_dogstatsd_amd64	-77.65 KiB (0.19% reduction, +9.15% of buffer)	39.082 → 39.006 → 39.910
✅	docker_dogstatsd_arm64	-77.63 KiB (0.20% reduction, +7.48% of buffer)	37.257 → 37.181 → 38.270

22 successful checks with minimal change (< 2 KiB)

	Quality gate	Current Size
✅	agent_heroku_amd64	309.997 MiB
✅	agent_msi	615.199 MiB
✅	agent_rpm_arm64	724.688 MiB
✅	agent_rpm_arm64_fips	684.257 MiB
✅	agent_suse_arm64	724.688 MiB
✅	agent_suse_arm64_fips	684.257 MiB
✅	docker_agent_arm64	809.424 MiB
✅	docker_agent_jmx_arm64	988.974 MiB
✅	docker_cluster_agent_arm64	221.440 MiB
✅	docker_cws_instrumentation_amd64	7.447 MiB
✅	docker_cws_instrumentation_arm64	6.877 MiB
✅	docker_host_profiler_amd64	304.564 MiB
✅	docker_host_profiler_arm64	315.680 MiB
✅	dogstatsd_deb_amd64	29.741 MiB
✅	dogstatsd_deb_arm64	27.800 MiB
✅	dogstatsd_rpm_amd64	29.741 MiB
✅	dogstatsd_suse_amd64	29.741 MiB
✅	iot_agent_deb_amd64	45.566 MiB
✅	iot_agent_deb_arm64	42.296 MiB
✅	iot_agent_deb_armhf	43.081 MiB
✅	iot_agent_rpm_amd64	45.566 MiB
✅	iot_agent_suse_amd64	45.565 MiB

[github-actions]

@codex review

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: b92303aaa4

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]

Commit the Bazel module lock update

bazel/codereview_guideline.md requires MODULE.bazel.lock to be committed whenever a module extension invocation changes. This adds a new go_deps repo exposure for the new Go module, but the commit leaves MODULE.bazel.lock untouched, so Bazel module resolution can run with stale lock state; please run bazel mod tidy/the repo’s Bazel dependency update flow and commit the lockfile change.

Useful? React with 👍 / 👎.

[chatgpt-codex-connector]

Regenerate the third-party license manifest

Adding this Go module also changes go.sum, but LICENSE-3rdparty.csv was not updated. The GitLab lint_licenses job runs dda inv -- -e lint-licenses, and tasks/go.py compares that CSV against the vendored dependency set from go.sum, so this dependency addition will leave the license manifest stale until dda inv generate-licenses is run and committed.

Useful? React with 👍 / 👎.

[cit-pr-commenter-54b7da]

Regression Detector

Regression Detector Results

Metrics dashboard
Target profiles
Run ID: 8b0df421-b7b5-4c79-8641-1335b53f82fa

Baseline: aced8ce
Comparison: 5acd305
Diff

Optimization Goals: ✅ No significant changes detected

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	quality_gate_idle	memory utilization	+0.15	[+0.08, +0.21]	1	Logs bounds checks dashboard
➖	quality_gate_idle_all_features	memory utilization	+0.11	[+0.07, +0.15]	1	Logs bounds checks dashboard
➖	quality_gate_logs	% cpu utilization	+0.06	[-1.02, +1.14]	1	Logs bounds checks dashboard
➖	quality_gate_metrics_logs	memory utilization	-0.07	[-0.32, +0.18]	1	Logs bounds checks dashboard

Bounds Checks: ✅ Passed

perf	experiment	bounds_check_name	replicates_passed	observed_value	links
✅	quality_gate_idle	intake_connections	10/10	3 ≤ 4	bounds checks dashboard
✅	quality_gate_idle	memory_usage	10/10	143.56MiB ≤ 154MiB	bounds checks dashboard
✅	quality_gate_idle	total_bytes_received	10/10	577.95KiB ≤ 819.20KiB	bounds checks dashboard
✅	quality_gate_idle_all_features	intake_connections	10/10	3 ≤ 4	bounds checks dashboard
✅	quality_gate_idle_all_features	memory_usage	10/10	479.99MiB ≤ 495MiB	bounds checks dashboard
✅	quality_gate_idle_all_features	total_bytes_received	10/10	0.89MiB ≤ 1.25MiB	bounds checks dashboard
✅	quality_gate_logs	intake_connections	10/10	4 ≤ 6	bounds checks dashboard
✅	quality_gate_logs	memory_usage	10/10	180.87MiB ≤ 195MiB	bounds checks dashboard
✅	quality_gate_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_logs	total_bytes_received	10/10	264.21MiB ≤ 292MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	cpu_usage	10/10	325.24 ≤ 2000	bounds checks dashboard
✅	quality_gate_metrics_logs	intake_connections	10/10	3 ≤ 6	bounds checks dashboard
✅	quality_gate_metrics_logs	memory_usage	10/10	390.72MiB ≤ 430MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_metrics_logs	total_bytes_received	10/10	0.86GiB ≤ 1.04GiB	bounds checks dashboard

Explanation

Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%

Performance changes are noted in the perf column of each table:

• ✅ = significantly better comparison variant performance
• ❌ = significantly worse comparison variant performance
• ➖ = no significant change in performance

A regression test is an A/B test of target performance in a repeatable rig, where "performance" is measured as "comparison variant minus baseline variant" for an optimization goal (e.g., ingress throughput). Due to intrinsic variability in measuring that goal, we can only estimate its mean value for each experiment; we report uncertainty in that value as a 90.00% confidence interval denoted "Δ mean % CI".

For each experiment, we decide whether a change in performance is a "regression" -- a change worth investigating further -- if all of the following criteria are true:

1. Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.

2. Its 90.00% confidence interval "Δ mean % CI" does not contain zero, indicating that if our statistical model is accurate, there is at least a 90.00% chance there is a difference in performance between baseline and comparison variants.

3. Its configuration does not mark it "erratic".

Replicate Execution Details

We run multiple replicates for each experiment/variant. However, we allow replicates to be automatically retried if there are any failures, up to 8 times, at which point the replicate is marked dead and we are unable to run analysis for the entire experiment. We call each of these attempts at running replicates a replicate execution. This section lists all replicate executions that failed due to the target crashing or being oom killed.

Note: In the below tables we bucket failures by experiment, variant, and failure type. For each of these buckets we list out the replicate indexes that failed with an annotation signifying how many times said replicate failed with the given failure mode. In the below example the baseline variant of the experiment named experiment_with_failures had two replicates that failed by oom kills. Replicate 0, which failed 8 executions, and replicate 1 which failed 6 executions, all with the same failure mode.

Experiment	Variant	Replicates	Failure	Logs	Debug Dashboard
experiment_with_failures	baseline	0 (x8) 1 (x6)	Oom killed		Debug Dashboard

The debug dashboard links will take you to a debugging dashboard specifically designed to investigate replicate execution failures.

❌ Retried Profiling Replicate Execution Failures (ddprof)

Note: Profiling replicas may still be executing. See the debug dashboard for up to date status.

Experiment	Variant	Replicates	Failure	Debug Dashboard
quality_gate_idle	baseline	10	Oom killed	Debug Dashboard
quality_gate_idle	comparison	10	Oom killed	Debug Dashboard
quality_gate_idle_all_features	baseline	10	Oom killed	Debug Dashboard
quality_gate_idle_all_features	comparison	10	Oom killed	Debug Dashboard
quality_gate_logs	baseline	10	Oom killed	Debug Dashboard
quality_gate_logs	comparison	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	baseline	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	comparison	10	Oom killed	Debug Dashboard

CI Pass/Fail Decision

✅ Passed. All Quality Gates passed.

• quality_gate_metrics_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check intake_connections: 10/10 replicas passed. Gate passed.

[github-actions]

@codex review

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 5acd3052e4

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]

Commit the updated Bazel lockfile

Adding com_github_datadog_dd_policy_engine_go changes the go_deps Bzlmod extension output, but MODULE.bazel.lock was not updated and still contains no entry for this repo (checked with rg com_github_datadog_dd_policy_engine MODULE.bazel.lock). The Bazel review guideline in bazel/codereview_guideline.md requires committing the refreshed lockfile for module-extension changes; otherwise Bazel lockfile-mode checks/reproducible fetches can fail before this new dependency is available. Please run the Bazel dependency refresh and commit the resulting MODULE.bazel.lock update.

Useful? React with 👍 / 👎.