-
Sub-task
-
Resolution: Fixed
-
P4
-
18
-
b26
| Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
|---|---|---|---|---|---|---|
| JDK-8299799 | 17.0.7 | Thomas Stuefe | P4 | Resolved | Fixed | b01 |
As laid out in JDK-8275301, it makes sense to let NMT do buffer overflow checking.
- it would give us C-heap overflow checking in release builds
- the additional costs are neglectable
- NMT needs intact headers anyway. Faced with buffer overwrites, today it may crash or cause other errors
- it is a preparation for future code removal (the memory guarding done in debug only in os::malloc() and friends, and possibly the guarding done with CheckJNICalls)
- it would give us C-heap overflow checking in release builds
- the additional costs are neglectable
- NMT needs intact headers anyway. Faced with buffer overwrites, today it may crash or cause other errors
- it is a preparation for future code removal (the memory guarding done in debug only in os::malloc() and friends, and possibly the guarding done with CheckJNICalls)
- backported by
-
-
- Resolved
-
- relates to
-
JDK-8298003 NMT: fatal block printout does not show the correct corruption address
-
- Resolved
-
-
-
- Resolved
-
- links to
-
Commit
openjdk/jdk17u-dev/6f115029
-
Commit
openjdk/jdk/cf7adae6
-
Review
openjdk/jdk17u-dev/842
-
Review
openjdk/jdk/5952
(2 links to)