-
Bug
-
Resolution: Fixed
-
P4
-
11, 17
-
b04
Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
---|---|---|---|---|---|---|
JDK-8260987 | 16.0.2 | Tobias Hartmann | P4 | Resolved | Fixed | b01 |
JDK-8260894 | 16.0.1 | Tobias Hartmann | P4 | Resolved | Fixed | b04 |
JDK-8259461 | 11.0.11 | Christian Hagedorn | P4 | Resolved | Fixed | b01 |
Backporting JDK-8257513 to JDK 11u triggered the following build failure:
[2021-01-04T09:54:55,894Z] workspace/open/src/hotspot/share/opto/compile.cpp: In member function 'bool Compile::ConstantTable::emit(CodeBuffer&) const':
[2021-01-04T09:54:55,894Z] workspace/open/src/hotspot/share/opto/compile.cpp:4136:83: error: 'last_addr' may be used uninitialized in this function [-Werror=maybe-uninitialized]
[2021-01-04T09:54:55,894Z] address new_constant_addr = last_addr - ((n->outcnt() - 1) * sizeof(address));
Not sure why this is not a problem in mainline. Maybe we build with different flags.
[2021-01-04T09:54:55,894Z] workspace/open/src/hotspot/share/opto/compile.cpp: In member function 'bool Compile::ConstantTable::emit(CodeBuffer&) const':
[2021-01-04T09:54:55,894Z] workspace/open/src/hotspot/share/opto/compile.cpp:4136:83: error: 'last_addr' may be used uninitialized in this function [-Werror=maybe-uninitialized]
[2021-01-04T09:54:55,894Z] address new_constant_addr = last_addr - ((n->outcnt() - 1) * sizeof(address));
Not sure why this is not a problem in mainline. Maybe we build with different flags.
- backported by
-
JDK-8259461 Uninitialized variable after JDK-8257513
-
- Resolved
-
-
JDK-8260894 Uninitialized variable after JDK-8257513
-
- Resolved
-
-
JDK-8260987 Uninitialized variable after JDK-8257513
-
- Resolved
-
- relates to
-
JDK-8257513 C2: assert((constant_addr - _masm.code()->consts()->start()) == con.offset())
-
- Resolved
-
- links to
-
Commit openjdk/jdk16u/c059d743
-
Commit openjdk/jdk/9f151649
-
Review openjdk/jdk16u/18
-
Review openjdk/jdk/1930
(3 links to)