Hongshi/mac sw inventory mem

[guohdd]

What does this PR do?

This PR addresses the issue of excessive memory footprint caused by caching in software inventory operation on MacOS.

Motivation

WINA-2530

Describe how you validated your changes

Collected multiple data sets of software inventory before and after this solution, confirm memory footprint stays at reasonable level with the solution. Ensure all CI pipelines pass.

Sample results below:

Steady-state RSS summary (KB)

• Baseline (released build)

  • agent: median 153,680, p95 153,697, max 153,712
  • system-probe: median 336,832, p95 336,832, max 336,832

• Improved (inventory enabled) (3 data sets)

  • agent: median 124,544, p95 126,896, max 126,896
  • system-probe: median 85,024, p95 85,232, max 85,232

• Improved (inventory disabled) (2 data sets)

  • agent: median 129,728, p95 129,904, max 129,904
  • system-probe: median 87,648, p95 87,776, max 87,776

Median delta vs baseline

• Improved (inventory enabled)
  • agent: -18.96%
  • system-probe: -74.76%

Bottom line

• system-probe steady footprint: from ~337 MB down to ~85–88 MB
• Enabling/disabling software inventory in the improved build changes memory only slightly compared to baseline reduction magnitude

Additional Notes

[agent-platform-auto-pr]

Files inventory check summary

File checks results against ancestor f0e9c8ed:

Results for datadog-agent_7.79.0~devel.git.294.ce8855b.pipeline.105408004-1_amd64.deb:

No change detected

[agent-platform-auto-pr]

Static quality checks

✅ Please find below the results from static quality gates
Comparison made with ancestor f0e9c8e
📊 Static Quality Gates Dashboard
🔗 SQG Job

30 successful checks with minimal change (< 2 KiB)

	Quality gate	Current Size
✅	agent_deb_amd64	753.050 MiB
✅	agent_deb_amd64_fips	710.005 MiB
✅	agent_heroku_amd64	313.310 MiB
✅	agent_rpm_amd64	753.034 MiB
✅	agent_rpm_amd64_fips	709.989 MiB
✅	agent_rpm_arm64	731.456 MiB
✅	agent_rpm_arm64_fips	691.428 MiB
✅	agent_suse_amd64	753.034 MiB
✅	agent_suse_amd64_fips	709.989 MiB
✅	agent_suse_arm64	731.456 MiB
✅	agent_suse_arm64_fips	691.428 MiB
✅	docker_agent_amd64	813.353 MiB
✅	docker_agent_arm64	816.545 MiB
✅	docker_agent_jmx_amd64	1004.268 MiB
✅	docker_agent_jmx_arm64	996.239 MiB
✅	docker_cluster_agent_amd64	203.937 MiB
✅	docker_cluster_agent_arm64	218.357 MiB
✅	docker_cws_instrumentation_amd64	7.142 MiB
✅	docker_cws_instrumentation_arm64	6.689 MiB
✅	docker_dogstatsd_amd64	39.234 MiB
✅	docker_dogstatsd_arm64	37.445 MiB
✅	dogstatsd_deb_amd64	29.878 MiB
✅	dogstatsd_deb_arm64	28.030 MiB
✅	dogstatsd_rpm_amd64	29.878 MiB
✅	dogstatsd_suse_amd64	29.878 MiB
✅	iot_agent_deb_amd64	43.285 MiB
✅	iot_agent_deb_arm64	40.332 MiB
✅	iot_agent_deb_armhf	41.080 MiB
✅	iot_agent_rpm_amd64	43.286 MiB
✅	iot_agent_suse_amd64	43.286 MiB

On-wire sizes (compressed)

	Quality gate	Change	Size (prev → curr → max)
✅	agent_deb_amd64	-14.25 KiB (0.01% reduction)	174.786 → 174.772 → 178.360
✅	agent_deb_amd64_fips	+22.53 KiB (0.01% increase)	165.361 → 165.383 → 172.790
✅	agent_heroku_amd64	neutral	74.998 MiB → 79.970
✅	agent_rpm_amd64	-13.96 KiB (0.01% reduction)	177.633 → 177.620 → 181.830
✅	agent_rpm_amd64_fips	-5.49 KiB (0.00% reduction)	167.645 → 167.639 → 173.370
✅	agent_rpm_arm64	-6.01 KiB (0.00% reduction)	159.550 → 159.544 → 163.060
✅	agent_rpm_arm64_fips	+20.41 KiB (0.01% increase)	151.368 → 151.387 → 156.170
✅	agent_suse_amd64	-13.96 KiB (0.01% reduction)	177.633 → 177.620 → 181.830
✅	agent_suse_amd64_fips	-5.49 KiB (0.00% reduction)	167.645 → 167.639 → 173.370
✅	agent_suse_arm64	-6.01 KiB (0.00% reduction)	159.550 → 159.544 → 163.060
✅	agent_suse_arm64_fips	+20.41 KiB (0.01% increase)	151.368 → 151.387 → 156.170
✅	docker_agent_amd64	neutral	268.200 MiB → 272.480
✅	docker_agent_arm64	neutral	255.380 MiB → 261.060
✅	docker_agent_jmx_amd64	-2.47 KiB (0.00% reduction)	336.856 → 336.854 → 341.100
✅	docker_agent_jmx_arm64	neutral	320.023 MiB → 325.620
✅	docker_cluster_agent_amd64	neutral	71.374 MiB → 72.920
✅	docker_cluster_agent_arm64	neutral	67.011 MiB → 68.220
✅	docker_cws_instrumentation_amd64	neutral	2.999 MiB → 3.330
✅	docker_cws_instrumentation_arm64	neutral	2.729 MiB → 3.090
✅	docker_dogstatsd_amd64	neutral	15.173 MiB → 15.820
✅	docker_dogstatsd_arm64	neutral	14.488 MiB → 14.830
✅	dogstatsd_deb_amd64	neutral	7.894 MiB → 8.790
✅	dogstatsd_deb_arm64	neutral	6.777 MiB → 7.710
✅	dogstatsd_rpm_amd64	neutral	7.905 MiB → 8.800
✅	dogstatsd_suse_amd64	neutral	7.905 MiB → 8.800
✅	iot_agent_deb_amd64	+2.98 KiB (0.03% increase)	11.403 → 11.406 → 12.040
✅	iot_agent_deb_arm64	neutral	9.705 MiB → 10.450
✅	iot_agent_deb_armhf	neutral	9.941 MiB → 10.620
✅	iot_agent_rpm_amd64	-3.1 KiB (0.03% reduction)	11.422 → 11.418 → 12.060
✅	iot_agent_suse_amd64	-3.1 KiB (0.03% reduction)	11.422 → 11.418 → 12.060

[cit-pr-commenter-54b7da]

Regression Detector

Regression Detector Results

Metrics dashboard
Target profiles
Run ID: e865d6e4-740f-4dab-892a-15ebe98a357a

Baseline: f0e9c8e
Comparison: ce8855b
Diff

Optimization Goals: ✅ No significant changes detected

Experiments ignored for regressions

Regressions in experiments with settings containing erratic: true are ignored.

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	docker_containers_cpu	% cpu utilization	-1.77	[-4.72, +1.18]	1	Logs

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	quality_gate_logs	% cpu utilization	+1.03	[-0.54, +2.60]	1	Logs bounds checks dashboard
➖	quality_gate_metrics_logs	memory utilization	+0.47	[+0.23, +0.71]	1	Logs bounds checks dashboard
➖	file_tree	memory utilization	+0.20	[+0.15, +0.26]	1	Logs
➖	quality_gate_idle_all_features	memory utilization	+0.13	[+0.09, +0.17]	1	Logs bounds checks dashboard
➖	otlp_ingest_logs	memory utilization	+0.09	[-0.00, +0.19]	1	Logs
➖	file_to_blackhole_0ms_latency	egress throughput	+0.06	[-0.45, +0.56]	1	Logs
➖	file_to_blackhole_1000ms_latency	egress throughput	+0.05	[-0.38, +0.49]	1	Logs
➖	file_to_blackhole_100ms_latency	egress throughput	+0.01	[-0.07, +0.10]	1	Logs
➖	tcp_dd_logs_filter_exclude	ingress throughput	-0.00	[-0.11, +0.11]	1	Logs
➖	uds_dogstatsd_to_api_v3	ingress throughput	-0.00	[-0.20, +0.19]	1	Logs
➖	uds_dogstatsd_to_api	ingress throughput	-0.01	[-0.21, +0.19]	1	Logs
➖	file_to_blackhole_500ms_latency	egress throughput	-0.01	[-0.41, +0.39]	1	Logs
➖	quality_gate_idle	memory utilization	-0.05	[-0.10, +0.00]	1	Logs bounds checks dashboard
➖	ddot_metrics_sum_cumulativetodelta_exporter	memory utilization	-0.09	[-0.31, +0.13]	1	Logs
➖	uds_dogstatsd_20mb_12k_contexts_20_senders	memory utilization	-0.13	[-0.19, -0.08]	1	Logs
➖	ddot_metrics_sum_delta	memory utilization	-0.27	[-0.44, -0.09]	1	Logs
➖	docker_containers_memory	memory utilization	-0.27	[-0.34, -0.20]	1	Logs
➖	otlp_ingest_metrics	memory utilization	-0.34	[-0.50, -0.17]	1	Logs
➖	ddot_logs	memory utilization	-0.52	[-0.59, -0.45]	1	Logs
➖	ddot_metrics_sum_cumulative	memory utilization	-0.62	[-0.76, -0.48]	1	Logs
➖	ddot_metrics	memory utilization	-0.66	[-0.84, -0.49]	1	Logs
➖	tcp_syslog_to_blackhole	ingress throughput	-0.93	[-1.09, -0.77]	1	Logs
➖	docker_containers_cpu	% cpu utilization	-1.77	[-4.72, +1.18]	1	Logs

Bounds Checks: ✅ Passed

perf	experiment	bounds_check_name	replicates_passed	observed_value	links
✅	docker_containers_cpu	simple_check_run	10/10	706 ≥ 26
✅	docker_containers_memory	memory_usage	10/10	271.58MiB ≤ 370MiB
✅	docker_containers_memory	simple_check_run	10/10	704 ≥ 26
✅	file_to_blackhole_0ms_latency	memory_usage	10/10	0.20GiB ≤ 1.20GiB
✅	file_to_blackhole_0ms_latency	missed_bytes	10/10	0B = 0B
✅	file_to_blackhole_1000ms_latency	memory_usage	10/10	0.23GiB ≤ 1.20GiB
✅	file_to_blackhole_1000ms_latency	missed_bytes	10/10	0B = 0B
✅	file_to_blackhole_100ms_latency	memory_usage	10/10	0.20GiB ≤ 1.20GiB
✅	file_to_blackhole_100ms_latency	missed_bytes	10/10	0B = 0B
✅	file_to_blackhole_500ms_latency	memory_usage	10/10	0.21GiB ≤ 1.20GiB
✅	file_to_blackhole_500ms_latency	missed_bytes	10/10	0B = 0B
✅	quality_gate_idle	intake_connections	10/10	3 = 3	bounds checks dashboard
✅	quality_gate_idle	memory_usage	10/10	174.52MiB ≤ 175MiB	bounds checks dashboard
✅	quality_gate_idle_all_features	intake_connections	10/10	3 = 3	bounds checks dashboard
✅	quality_gate_idle_all_features	memory_usage	10/10	494.23MiB ≤ 550MiB	bounds checks dashboard
✅	quality_gate_logs	intake_connections	10/10	4 ≤ 6	bounds checks dashboard
✅	quality_gate_logs	memory_usage	10/10	204.71MiB ≤ 220MiB	bounds checks dashboard
✅	quality_gate_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_metrics_logs	cpu_usage	10/10	348.60 ≤ 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	412.83MiB ≤ 475MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	missed_bytes	10/10	0B = 0B	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".

CI Pass/Fail Decision

✅ Passed. All Quality Gates passed.

• quality_gate_idle_all_features, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check missed_bytes: 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_logs, 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_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check intake_connections: 10/10 replicas passed. Gate passed.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 4d10b7185d

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]

Prevent cache eviction from undoing prefetch results

Collect prefetches summaries for all receipts and then does a second full pass that calls cache.get again, but this eviction path keeps only 512 entries, so any host with more than 512 receipts will evict at least N-512 prefetched items before they are consumed. In that case the second pass re-runs pkgutil --files synchronously for the evicted receipts, causing a large command-count and latency regression (often close to doubling work) on long-lived macOS systems with many installer receipts.

Useful? React with 👍 / 👎.

[guohdd]

The pre-defined limit of 512 is about pkg receipts. It could be a rare case that a system has more than 512 pkg receipts. We may need to see some numbers to further look into the concern.