Bug 1718838 - Don't call pthread_equal() on possibly-invalid thread IDs

[larseggert]

lock/mon owner fields can be zero-initialized; pthread_equal() on that is UB. Add pt_pthread_equal(): fall back to == only for the invalid sentinel, else delegate to real pthread_equal(). Used wherever pthread_equal() compared a possibly-unset owner, in ptsynch.c and ptthread.c.

https://bugzilla.mozilla.org/show_bug.cgi?id=1718838

From NetBSD pkgsrc.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 0% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 48.91%. Comparing base (6bac64c) to head (6d85421).
⚠️ Report is 23 commits behind head on main.

Files with missing lines	Patch %	Lines
pr/src/pthreads/ptsynch.c	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #66      +/-   ##
==========================================
+ Coverage   42.20%   48.91%   +6.70%     
==========================================
  Files          70       93      +23     
  Lines       12071    17828    +5757     
  Branches     2261     2261              
==========================================
+ Hits         5095     8720    +3625     
- Misses       6126     8687    +2561     
+ Partials      850      421     -429     

Flag	Coverage Δ
freebsd	42.65% <0.00%> (+0.08%)	⬆️
macos-15-intel	42.70% <0.00%> (+0.01%)	⬆️
macos-latest	42.72% <0.00%> (+0.03%)	⬆️
netbsd	39.37% <0.00%> (+0.08%)	⬆️
openbsd	1.00% <0.00%> (ø)
solaris	39.77% <0.00%> (ø)
ubuntu-24.04-arm-clang	41.76% <0.00%> (-0.01%)	⬇️
ubuntu-latest-gcc	39.47% <0.00%> (-0.05%)	⬇️
ubuntu-latest-gcc-x86	39.49% <0.00%> (-0.01%)	⬇️
windows-arm64-cl	51.16% <ø> (?)
windows-x64-cl	51.46% <ø> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Copilot]

Copilot review overview

🟡 Changes recommended

Valid thread handles should retain POSIX comparison semantics, and the invalid-owner path needs regression coverage.

Get a fresh assessment by requesting another Copilot review.

Review effort: Balanced
Findings: 1 · 1

Open (2)

• Use pthread_equal() for valid pthread_t comparisons · New
• Add regression test for zero count on new monitors · New

What changed in this PR

Prevents undefined behavior when comparing uninitialized pthread owner IDs.

Changes:

• Adds a safe thread-ID comparison helper.
• Updates lock, monitor, and thread comparisons.

File	Description
pr/​src/​pthreads/​ptthread.c	Uses the new comparison helper.
pr/​src/​pthreads/​ptsynch.c	Adds the helper and updates owner comparisons.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[github-actions]

nspr4 backport check

Commit	Alone	In order	Files	ABI	Subject
6d854212	yes	yes	source	unchanged	Bug 1718838 - Don't call pthread_equal() on possibly-invalid thread IDs

Alone: cherry-picks onto nspr4 by itself. In order: cherry-picks onto nspr4 after the earlier commits, skipping any that conflict.

Verdict: backportable

[larseggert]

@sylvestre thanks, that really simplified the diff :-)