Commit e35c8af
revert: undo per-crate --test-threads=1 split for musl CI (#289)
## Summary
Reverts fa42ef9 ("ci: replace global
RUST_TEST_THREADS=1 with per-crate --test-threads=1 (#280)").
The change replaced the simple `RUST_TEST_THREADS=1` env var with
`cargo-nextest` installation and split test commands. This actually
**regressed** musl CI performance:
### CI Time Comparison (musl job)
| | Before (7a8ad8b) | After (fa42ef9) | Delta |
|---|---|---|---|
| **Test (musl)** | **4m 31s** | **12m 32s** | **+8m 01s (+177%)** |
The overhead comes from installing `cargo-nextest` and running two
separate `cargo` invocations instead of one.
This revert restores the global `RUST_TEST_THREADS=1` approach, bringing
the musl CI job back to ~4.5 minutes.
## Test plan
- [ ] Verify musl CI job passes
- [ ] Verify musl CI job completes in ~4-5 minutes instead of ~12
minutes
https://claude.ai/code/session_01LGzZqKECkrqipZfmmU9L3U
Co-authored-by: Claude <noreply@anthropic.com>1 parent b4ba22c commit e35c8af
1 file changed
+4
-8
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
149 | 149 | | |
150 | 150 | | |
151 | 151 | | |
| 152 | + | |
| 153 | + | |
| 154 | + | |
152 | 155 | | |
153 | 156 | | |
154 | 157 | | |
| |||
172 | 175 | | |
173 | 176 | | |
174 | 177 | | |
175 | | - | |
176 | | - | |
177 | | - | |
178 | | - | |
179 | | - | |
180 | | - | |
181 | | - | |
182 | | - | |
| 178 | + | |
183 | 179 | | |
184 | 180 | | |
185 | 181 | | |
| |||
0 commit comments