[Dynamic Instrumentation] refactor loadFunctionDefinitions to use one reader
#34594
Conversation
|
@grantseltzer I know it's not a priority now, but I wanted to get it down while it was still in my head. I think for particularly large DWARF objects, we're performing DFS which means we keep everything in memory until the traversal halts. If we switch this to BFS, that should allow the GC to cleanup objects and avoid potential OOM problems. It gets a lot more complicated further down the chain, so I limited the scope of this refactor to just the |
8c5f6c9 to
04f116b
Compare
|
[Fast Unit Tests Report] On pipeline 57403993 (CI Visibility). The following jobs did not run any unit tests: Jobs:
If you modified Go files and expected unit tests to run in these jobs, please double check the job logs. If you think tests should have been executed reach out to #agent-devx-help |
Test changes on VMUse this command from test-infra-definitions to manually test this PR changes on a VM: inv aws.create-vm --pipeline-id=57403993 --os-family=ubuntuNote: This applies to commit d44dae0 |
Uncompressed package size comparisonComparison with ancestor Diff per package
Decision✅ Passed |
Static quality checks ✅Please find below the results from static quality gates Successful checksInfo
|
Regression DetectorRegression Detector ResultsMetrics dashboard Baseline: 734ced8 Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | quality_gate_idle | memory utilization | +0.46 | [+0.42, +0.50] | 1 | Logs bounds checks dashboard |
| ➖ | file_to_blackhole_500ms_latency | egress throughput | +0.10 | [-0.68, +0.88] | 1 | Logs |
| ➖ | tcp_syslog_to_blackhole | ingress throughput | +0.07 | [+0.01, +0.13] | 1 | Logs |
| ➖ | quality_gate_idle_all_features | memory utilization | +0.02 | [-0.05, +0.08] | 1 | Logs bounds checks dashboard |
| ➖ | file_to_blackhole_100ms_latency | egress throughput | +0.01 | [-0.67, +0.69] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http1 | egress throughput | +0.01 | [-0.78, +0.80] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api | ingress throughput | +0.01 | [-0.30, +0.31] | 1 | Logs |
| ➖ | file_to_blackhole_300ms_latency | egress throughput | +0.00 | [-0.63, +0.63] | 1 | Logs |
| ➖ | tcp_dd_logs_filter_exclude | ingress throughput | -0.00 | [-0.02, +0.02] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency_linear_load | egress throughput | -0.00 | [-0.48, +0.47] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency | egress throughput | -0.01 | [-0.84, +0.82] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http2 | egress throughput | -0.01 | [-0.79, +0.77] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api_cpu | % cpu utilization | -0.11 | [-0.99, +0.76] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency | egress throughput | -0.52 | [-1.30, +0.25] | 1 | Logs |
| ➖ | file_tree | memory utilization | -0.53 | [-0.59, -0.48] | 1 | Logs |
| ➖ | quality_gate_logs | % cpu utilization | -0.84 | [-3.72, +2.04] | 1 | Logs |
Bounds Checks: ✅ Passed
| perf | experiment | bounds_check_name | replicates_passed | links |
|---|---|---|---|---|
| ✅ | file_to_blackhole_0ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http1 | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http1 | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http2 | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http2 | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency_linear_load | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | memory_usage | 10/10 | |
| ✅ | quality_gate_idle | intake_connections | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | intake_connections | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_logs | intake_connections | 10/10 | |
| ✅ | quality_gate_logs | lost_bytes | 10/10 | |
| ✅ | quality_gate_logs | memory_usage | 10/10 |
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:
-
Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.
-
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.
-
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_idle, 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_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check lost_bytes: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
04f116b to
d44dae0
Compare
What does this PR do?
This pull request includes several changes to the
pkg/dynamicinstrumentation/diconfig/dwarf.gofile to streamline the process of reading and expanding type data from DWARF debug information. The modifications focus on optimizing the use of the DWARF reader and improving the handling of type offsets.Improvements to DWARF data reading and type expansion:
entryReaderandtypeReaderand using a singlereaderinstead. [1] [2]offsetsto collect offsets of type entries, which are then processed in a separate loop to expand type data. [1] [2]expandTypeDatafunction to accept areaderparameter, allowing it to use the same reader instance for seeking and reading type entries.resolveTypedefToRealTypefunction calls to use the consolidatedreaderinstead of the removedtypeReader.Motivation
As part of incident-35560, we noticed that the
loadFunctionDefinitionshad a higher start-up cost than expected. This reduces the number of open readers and is a first step to a smaller memory footprint.Describe how you validated your changes
unit / e2e
Possible Drawbacks / Trade-offs
Additional Notes