[araujogui]

pop_async_context() called shrink_to_fit() on every pop. The stack is usually empty once a callback returns, so libc++ and MSVC free the deque's storage and the next push allocates it again. That happens several times per HTTP request.

The shrink used to run only when the stack was large and mostly unused. That check was dropped in #56457.

Measured on macOS (Apple M2, libc++). libstdc++ skips the free in this case, so the gain on Linux is likely small.

benchmark/http/simple.js

                                                                                              confidence   improvement   accuracy (*)    (**)   (***)
http/simple.js duration=3 chunkedEnc=0 c=50 chunks=1 len=4 type='bytes' benchmarker='wrk'     ***             +4.44 %   ±2.15%  ±2.92%  ±3.93%
http/simple.js duration=3 chunkedEnc=1 c=50 chunks=1 len=4 type='bytes' benchmarker='wrk'                     +1.39 %   ±1.79%  ±2.44%  ±3.31%
http/simple.js duration=3 chunkedEnc=0 c=50 chunks=4 len=4 type='bytes' benchmarker='wrk'       *             +2.25 %   ±1.93%  ±2.63%  ±3.54%
http/simple.js duration=3 chunkedEnc=1 c=50 chunks=4 len=4 type='bytes' benchmarker='wrk'      **             +2.26 %   ±1.40%  ±1.91%  ±2.57%
http/simple.js duration=3 chunkedEnc=0 c=50 chunks=1 len=1024 type='bytes' benchmarker='wrk'  ***             +2.46 %   ±0.94%  ±1.29%  ±1.75%
http/simple.js duration=3 chunkedEnc=1 c=50 chunks=1 len=1024 type='bytes' benchmarker='wrk'   **             +1.92 %   ±1.39%  ±1.89%  ±2.56%
http/simple.js duration=3 chunkedEnc=0 c=50 chunks=4 len=1024 type='bytes' benchmarker='wrk'  ***             +1.93 %   ±0.80%  ±1.09%  ±1.46%
http/simple.js duration=3 chunkedEnc=1 c=50 chunks=4 len=1024 type='bytes' benchmarker='wrk'  ***             +2.91 %   ±1.19%  ±1.62%  ±2.19%

Be aware that when doing many comparisons the risk of a false-positive
result increases. In this case, there are 8 comparisons, you can thus
expect the following amount of false-positive results:
  0.40 false positives, when considering a   5% risk acceptance (*, **, ***),
  0.08 false positives, when considering a   1% risk acceptance (**, ***),
  0.01 false positives, when considering a 0.1% risk acceptance (***)

Refs: #56457

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.

[gurgunday]

FYI it's a duplicate of #66344

[araujogui]

FYI it's a duplicate of #66344

ty I hadn't seen that PR, closing this as duplicate.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.74%. Comparing base (97aed16) to head (8aa938a).
⚠️ Report is 109 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #67362      +/-   ##
==========================================
+ Coverage   90.67%   90.74%   +0.06%     
==========================================
  Files         797      798       +1     
  Lines      278773   280016    +1243     
  Branches    53607    53857     +250     
==========================================
+ Hits       252778   254092    +1314     
+ Misses      16356    16219     -137     
- Partials     9639     9705      +66     

Files with missing lines	Coverage Δ
src/env.cc	84.36% <ø> (-0.02%)	⬇️

... and 107 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.