Details
Details
- Reviewers
dcoughlin NoQ - Commits
- rG6794aa702a5e: [analyzer] [RetainCountChecker] Bugfix: in non-OSObject-mode, do not track CXX…
rL353227: [analyzer] [RetainCountChecker] Bugfix: in non-OSObject-mode, do not track CXX…
rC353227: [analyzer] [RetainCountChecker] Bugfix: in non-OSObject-mode, do not track CXX…
Diff Detail
Diff Detail
- Repository
- rC Clang
Event Timeline
clang/lib/Analysis/RetainSummaryManager.cpp | ||
---|---|---|
503 ↗ | (On Diff #185392) | So if an OSObject-related method deals with NS or CF objects, we wouldn't escape it anymore? I thing there might exist such code somewhere in between kernel and userland. |
clang/test/Analysis/retain-release.mm | ||
489 ↗ | (On Diff #185392) | I enjoy namespacing my generic names, this way i can use generic names without being afraid to collide with someone else's code (cf. the test above). |