fix: cel env after cache lookup by adityathebe · Pull Request #161 · flanksource/gomplate

adityathebe · 2026-06-23T10:51:09Z

Summary by CodeRabbit

Tests
- Added benchmarks for expression evaluation performance testing across different configuration sizes.
Refactor
- Optimized expression caching to skip unnecessary environment setup on cache hits.
Chores
- Enhanced CI/CD benchmark workflow to run expanded performance test suite on pull requests.

RunExpressionContext built the full CEL env option set (GetCelEnv, including kubernetes.Library() and one cel.Variable per env key) on every call before consulting the compiled-program cache, then discarded it on cache hits, since cel.NewEnv only runs to compile a new program. This made GetCelEnv (notably kubernetes.Library()) the dominant lifetime allocator on the CEL path: in a production config-db heap profile it accounted for ~38% of alloc_space, called on every evaluation. Move the cache lookup ahead of GetCelEnv and construct env options only inside the cache-miss branch where cel.NewEnv actually consumes them. Cache-hit steady state now skips GetCelEnv entirely. Behavior is unchanged: the cached cel.Program already carries its variable bindings and Eval still receives the serialized data.

Add BenchmarkRunExpressionContext (cacheHit smallEnv/largeEnv) guarding the cache-hit steady state where GetCelEnv must not run, modeled on the production config-item env shape with no duty/config-db dependencies. Wire it into the benchmark workflow alongside BenchmarkSerialize: base-vs-head benchstat, a report comment posted to the PR, and the existing 5% regression gate. Copy the new bench file into the base worktree so both runs execute the same suite, only the implementation under test differing.

coderabbitai · 2026-06-23T10:51:25Z

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 08d5f45e-cb09-4c5c-997b-eacafd7d4c79

📥 Commits

Reviewing files that changed from the base of the PR and between d6bef8e and 23e1cff.

📒 Files selected for processing (3)

.github/workflows/benchmark.yml
run_expression_bench_test.go
template.go

Walkthrough

RunExpressionContext in template.go is refactored to defer CEL environment option construction (GetCelEnv, function wrapping, CelEnvs) into the cache-miss branch, avoiding allocations on cache hits. A new BenchmarkRunExpressionContext benchmark measures this hot path across small and large env shapes, and the CI benchmark workflow is updated to include it.

Changes

CEL Cache-Hit Optimization and Benchmarking

Layer / File(s)	Summary
Defer `envOptions` to cache-miss path `template.go`	`GetCelEnv`, function wrapping, and `CelEnvs` population are moved inside the `prg == nil` block so they are skipped on cache hits; comments document the deferral intent.
`BenchmarkRunExpressionContext` and CI wiring `run_expression_bench_test.go`, `.github/workflows/benchmark.yml`	New benchmark defines a constant CEL expression, runs small and large sub-benchmarks that warm the cache once then measure allocations per iteration; CI workflow copies the new benchmark file into the base worktree and expands both base and head `go test -bench` filters to include `BenchmarkRunExpressionContext`.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately reflects the main change: deferring CEL environment option construction until after cache lookup in RunExpressionContext, which is an optimization to avoid unnecessary work on cache hits.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

Create stacked PR
Commit on current branch

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/cel-env-after-cache-lookup

✨ Simplify code

Create PR with simplified code
Commit simplified code in branch fix/cel-env-after-cache-lookup

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

github-actions · 2026-06-23T10:51:29Z

Benchstat

Base: d6bef8ed6b708ce16cf9a1e721bd46168af35aab
Head: 23e1cff2d7f984b0bd4be5abf9f8dc80d5516aad

6 improvement(s)

Benchmark	Base	Head	Change	p-value
`RunExpressionContext/cacheHit/smallEnv-4`	8593.0	664.0	-92.27%	0.002
`RunExpressionContext/cacheHit/smallEnv-4`	187.00	20.00	-89.30%	0.002
`RunExpressionContext/cacheHit/smallEnv-4`	9.388µ	1.810µ	-80.72%	0.002
`RunExpressionContext/cacheHit/largeEnv-4`	10.235Ki	2.445Ki	-76.11%	0.002
`RunExpressionContext/cacheHit/largeEnv-4`	269.0	101.0	-62.45%	0.002
`RunExpressionContext/cacheHit/largeEnv-4`	18.60µ	10.85µ	-41.66%	0.002

Full benchstat output

goos: linux
goarch: amd64
pkg: github.com/flanksource/gomplate/v3
cpu: AMD EPYC 7763 64-Core Processor                
                                         │ bench-base.txt │           bench-head.txt           │
                                         │     sec/op     │   sec/op     vs base               │
RunExpressionContext/cacheHit/smallEnv-4     9.388µ ± 24%   1.810µ ± 1%  -80.72% (p=0.002 n=6)
RunExpressionContext/cacheHit/largeEnv-4     18.60µ ±  1%   10.85µ ± 8%  -41.66% (p=0.002 n=6)
Serialize/Size-10-4                          37.17µ ±  3%   36.68µ ± 0%        ~ (p=0.065 n=6)
Serialize/Size-100-4                         354.5µ ±  1%   351.3µ ± 2%        ~ (p=0.240 n=6)
Serialize/Size-1000-4                        3.521m ±  1%   3.529m ± 0%        ~ (p=0.485 n=6)
Serialize/Size-10000-4                       40.91m ±  1%   40.49m ± 2%        ~ (p=0.132 n=6)
Serialize_NoNativeTypes/Size-100-4           90.65µ ±  1%   91.53µ ± 2%        ~ (p=0.485 n=6)
Serialize_NoNativeTypes/Size-1000-4          921.4µ ±  1%   924.9µ ± 4%        ~ (p=0.093 n=6)
Serialize_NoNativeTypes/Size-10000-4         9.330m ±  1%   9.282m ± 0%        ~ (p=0.093 n=6)
geomean                                      399.3µ         312.5µ       -21.75%

                                         │ bench-base.txt │            bench-head.txt             │
                                         │      B/op      │     B/op      vs base                 │
RunExpressionContext/cacheHit/smallEnv-4      8593.0 ± 0%     664.0 ± 0%  -92.27% (p=0.002 n=6)
RunExpressionContext/cacheHit/largeEnv-4    10.235Ki ± 0%   2.445Ki ± 0%  -76.11% (p=0.002 n=6)
Serialize/Size-10-4                          11.06Ki ± 0%   11.06Ki ± 0%        ~ (p=1.000 n=6) ¹
Serialize/Size-100-4                         95.52Ki ± 0%   95.51Ki ± 0%        ~ (p=0.701 n=6)
Serialize/Size-1000-4                        952.7Ki ± 0%   952.7Ki ± 0%        ~ (p=0.732 n=6)
Serialize/Size-10000-4                       10.06Mi ± 1%   10.05Mi ± 0%        ~ (p=1.000 n=6)
Serialize_NoNativeTypes/Size-100-4           36.33Ki ± 0%   36.33Ki ± 0%        ~ (p=1.000 n=6)
Serialize_NoNativeTypes/Size-1000-4          365.6Ki ± 0%   365.6Ki ± 0%        ~ (p=0.545 n=6)
Serialize_NoNativeTypes/Size-10000-4         3.584Mi ± 0%   3.584Mi ± 0%        ~ (p=0.303 n=6)
geomean                                      152.0Ki        97.56Ki       -35.84%
¹ all samples are equal

                                         │ bench-base.txt │            bench-head.txt            │
                                         │   allocs/op    │  allocs/op   vs base                 │
RunExpressionContext/cacheHit/smallEnv-4      187.00 ± 0%    20.00 ± 0%  -89.30% (p=0.002 n=6)
RunExpressionContext/cacheHit/largeEnv-4       269.0 ± 0%    101.0 ± 0%  -62.45% (p=0.002 n=6)
Serialize/Size-10-4                            268.0 ± 0%    268.0 ± 0%        ~ (p=1.000 n=6) ¹
Serialize/Size-100-4                          2.518k ± 0%   2.518k ± 0%        ~ (p=1.000 n=6) ¹
Serialize/Size-1000-4                         25.80k ± 0%   25.80k ± 0%        ~ (p=1.000 n=6)
Serialize/Size-10000-4                        265.1k ± 0%   265.0k ± 0%        ~ (p=1.000 n=6)
Serialize_NoNativeTypes/Size-100-4             911.0 ± 0%    911.0 ± 0%        ~ (p=1.000 n=6) ¹
Serialize_NoNativeTypes/Size-1000-4           9.758k ± 0%   9.758k ± 0%        ~ (p=1.000 n=6) ¹
Serialize_NoNativeTypes/Size-10000-4          100.1k ± 0%   100.1k ± 0%        ~ (p=0.197 n=6)
geomean                                       3.896k        2.725k       -30.04%
¹ all samples are equal

adityathebe added 2 commits June 23, 2026 15:38

github-actions Bot added the build label Jun 23, 2026

moshloop merged commit f34f0dc into main Jun 23, 2026
9 checks passed

moshloop deleted the fix/cel-env-after-cache-lookup branch June 23, 2026 13:57

coderabbitai Bot mentioned this pull request Jun 24, 2026

perf(cel): build static CEL env once and Extend per call #162

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix: cel env after cache lookup#161

fix: cel env after cache lookup#161
moshloop merged 2 commits into
mainfrom
fix/cel-env-after-cache-lookup

adityathebe commented Jun 23, 2026 •

edited by coderabbitai Bot

Loading

Uh oh!

coderabbitai Bot commented Jun 23, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Jun 23, 2026 •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

adityathebe commented Jun 23, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Uh oh!

coderabbitai Bot commented Jun 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Uh oh!

github-actions Bot commented Jun 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchstat

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

adityathebe commented Jun 23, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Jun 23, 2026 •

edited

Loading

github-actions Bot commented Jun 23, 2026 •

edited

Loading