| Summary: | Year 2038 bugs | ||
|---|---|---|---|
| Product: | [Technology] JGit | Reporter: | Shawn Pearce <sop> |
| Component: | JGit | Assignee: | Project Inbox <jgit.core-inbox> |
| Status: | ASSIGNED --- | QA Contact: | |
| Severity: | enhancement | ||
| Priority: | P3 | CC: | fcostasilva, matthias.sohn |
| Version: | unspecified | ||
| Target Milestone: | --- | ||
| Hardware: | All | ||
| OS: | All | ||
| See Also: |
https://git.eclipse.org/r/148014 https://git.eclipse.org/c/jgit/jgit.git/commit/?id=c130e5e708a34619a90474afd8f259931fc8e7c2 |
||
| Whiteboard: | |||
|
Description
Shawn Pearce
The problem with smudge racily clean algorithm was fixed in c98d97731b87417b196341fa63a50fffea4e123c. LStat : The ctime and mtime fields in class LStat are only 4 bytes wide, and are used as signed int. In 2038 these fields will roll over on us and start to report negative values. The problem with RevCommit hasn't been fixed. The code is pretty clear about this: ... int commitTime; // An int here for performance, overflows in 2038 New Gerrit change created: https://git.eclipse.org/r/148014 Gerrit change https://git.eclipse.org/r/148014 was merged to [stable-5.1]. Commit: http://git.eclipse.org/c/jgit/jgit.git/commit/?id=c130e5e708a34619a90474afd8f259931fc8e7c2 |