[EBPF] Generate complexity data in CI

[gjulianm]

What does this PR do?

Enables the generation of complexity data files in KMT CI. It also fixes some errors in the verifier log parser detected during testing of this PR.

Motivation

A future PR will leverage these files to report complexity changes in PRs.

Additional Notes

Possible Drawbacks / Trade-offs

Describe how to test/QA your changes

[pr-commenter]

Regression Detector

Regression Detector Results

Run ID: f84fba87-a0f7-4ad3-9303-af459216b0ac Metrics dashboard Target profiles

Baseline: 9551500
Comparison: b14de82

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

Significant changes in experiment optimization goals

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

perf	experiment	goal	Δ mean %	Δ mean % CI	links
❌	pycheck_1000_100byte_tags	% cpu utilization	+5.02	[+0.27, +9.78]	Logs

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	links
❌	pycheck_1000_100byte_tags	% cpu utilization	+5.02	[+0.27, +9.78]	Logs
➖	uds_dogstatsd_to_api_cpu	% cpu utilization	+0.92	[+0.04, +1.80]	Logs
➖	otel_to_otel_logs	ingress throughput	+0.76	[-0.06, +1.57]	Logs
➖	tcp_syslog_to_blackhole	ingress throughput	+0.05	[-12.79, +12.89]	Logs
➖	tcp_dd_logs_filter_exclude	ingress throughput	-0.00	[-0.01, +0.01]	Logs
➖	uds_dogstatsd_to_api	ingress throughput	-0.00	[-0.00, +0.00]	Logs
➖	idle	memory utilization	-0.07	[-0.10, -0.03]	Logs
➖	file_tree	memory utilization	-0.45	[-0.50, -0.40]	Logs
➖	basic_py_check	% cpu utilization	-1.28	[-3.88, +1.31]	Logs

Explanation

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".

[gjulianm]

/trigger-ci --variable RUN_ALL_BUILDS=true --variable RUN_KITCHEN_TESTS=true --variable RUN_E2E_TESTS=on --variable RUN_UNIT_TESTS=on --variable RUN_KMT_TESTS=on

[dd-devflow]

🚂 Gitlab pipeline started

Started pipeline #37571002

[pr-commenter]

Test changes on VM

Use this command from test-infra-definitions to manually test this PR changes on a VM:

inv create-vm --pipeline-id=37957316 --os-family=ubuntu

Note: This applies to commit b14de82

[val06]

reviewed

[val06]

the 1gb is accounted by the VerifierLog field of the Program struct?

in that case, maybe it is better to set explicitly the string to empty string, since the program parent struct is a held by pointer and it might cause the GC to skip freeing that memory (or delay the actual free operation)

[brycekahle]

You may need to do both: set the string to "" and call debug.FreeOSMemory()

[gjulianm]

I tested it with both ways and setting the string to "" was not enough to make the GC collect the memory, that's why I added debug.FreeOSMemory(). Just in case I changed the code to set the verifier log to an empty string to ensure it's freed, although I think that prog being scoped to just the loop is enough for the GC to mark that memory as unused.

[val06]

golang GC works a bit differently when it comes to pointers and heap memory

[gjulianm]

/merge

[dd-devflow]

🚂 MergeQueue: pull request added to the queue

The median merge time in main is 26m.

Use /merge -c to cancel this operation!