From 6e7ab529297ac3e8ef9110718e1eab688d4391ef Mon Sep 17 00:00:00 2001
From: "Adam Perry @ OpenAI" <anp@openai.com>
Date: Tue, 9 Jun 2026 12:48:04 -0700
Subject: [PATCH] [codex] Speed up local nextest runs (#26479)

## Why

`just test` currently uses the CI-oriented nextest profile, which
serializes app-server integration tests even on developer machines that
can run several safely. Bounded local parallelism substantially shortens
this common iteration loop without changing CI behavior.

Eight-worker experiments were faster, but keeping them reliable required
relaxing several test deadlines. Four workers for integration tests is a
solid tradeoff that speeds up local testing without needing to change
test logic.

## What changed

- Add a `local` nextest profile that inherits the existing defaults.
- Allow up to four app-server integration tests to run concurrently
under that profile.
- Make `just test` select the local profile on Unix and Windows.
- Keep the default CI profile serialized and leave all test deadlines
unchanged.

The tests use separate processes, randomized temporary `CODEX_HOME`
directories, and ephemeral ports. The remaining shared constraints are
system resources; each app-server also uses a multi-thread Tokio
runtime, and fuzzy-search tests can create additional worker threads, so
the local cap remains intentionally conservative.

## Performance and validation

All measurements below are warm, execution-only app-server runs with
nextest retries disabled.

On the current rebased branch, an AMD EPYC 7763 machine with 16 logical
CPUs and 62 GiB RAM completed three consecutive runs:

| Run | Nextest time | Wall time | Result |
| --- | ---: | ---: | --- |
| 1 | 142.941s | 145.17s | 836/836 passed |
| 2 | 143.402s | 145.59s | 836/836 passed |
| 3 | 142.870s | 145.08s | 836/836 passed |

The mean wall time was 145.28s. The slow-inventory, approval replay, and
zsh-fork tests all passed with their original deadlines.

Earlier measurements on the same Linux machine, before the suite grew,
showed the scaling that motivated the change:

| App-server concurrency | Nextest time | Result |
| --- | ---: | --- |
| 1 | 369.5s | 572/572 passed |
| 2 | 194.5s | 572/572 passed |
| 4 | 111.0s mean over 3 runs | 3/3 clean |

Four workers reduced that execution time by about 70%, a roughly 3.3x
speedup over serialization.
---
 codex-rs/.config/nextest.toml | 14 ++++++++++++++
 justfile                      |  4 ++--
 2 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/codex-rs/.config/nextest.toml b/codex-rs/.config/nextest.toml
index 01f4a98dc..f3f3ca260 100644
--- a/codex-rs/.config/nextest.toml
+++ b/codex-rs/.config/nextest.toml
@@ -8,12 +8,20 @@ retries = 1
 [profile.default.junit]
 path = "junit.xml"
 
+[profile.local]
+inherits = "default"
+
 [test-groups.app_server_protocol_codegen]
 max-threads = 1
 
 [test-groups.app_server_integration]
 max-threads = 1
 
+# Higher concurrency causes integration test timeouts under resource contention
+# on common developer machines.
+[test-groups.app_server_integration_local]
+max-threads = 4
+
 [test-groups.core_apply_patch_cli_integration]
 max-threads = 1
 
@@ -42,6 +50,12 @@ test-group = 'app_server_protocol_codegen'
 filter = 'package(codex-app-server) & kind(test)'
 test-group = 'app_server_integration'
 
+[[profile.local.overrides]]
+# Use up to four app-server subprocesses locally. The global nextest pool still
+# limits this to the machine's logical CPU count.
+filter = 'package(codex-app-server) & kind(test)'
+test-group = 'app_server_integration_local'
+
 [[profile.default.overrides]]
 # These tests exercise full Codex turns and apply_patch execution, and they are
 # sensitive to Windows runner process-startup stalls when many cases launch at once.
diff --git a/justfile b/justfile
index 3b993b767..fa32a19e7 100644
--- a/justfile
+++ b/justfile
@@ -75,11 +75,11 @@ install:
 # there should be no need to add `--all-features`.
 [unix]
 test *args:
-    RUST_MIN_STACK={{ rust_min_stack }} cargo nextest run --no-fail-fast "$@"
+    RUST_MIN_STACK={{ rust_min_stack }} NEXTEST_PROFILE=local cargo nextest run --no-fail-fast "$@"
 
 [windows]
 test *args:
-    $env:RUST_MIN_STACK = "{{ rust_min_stack }}"; cargo nextest run --no-fail-fast @($args | Select-Object -Skip 1)
+    $env:RUST_MIN_STACK = "{{ rust_min_stack }}"; $env:NEXTEST_PROFILE = "local"; cargo nextest run --no-fail-fast @($args | Select-Object -Skip 1)
 
 # Run from the repository root so scripts that resolve paths from `cwd` see
 # the same layout they use in GitHub Actions.