JDK-8254695 : G1: Next mark bitmap clear not cancelled after marking abort
- Type: Bug
- Component: hotspot
- Sub-Component: gc
- Affected Version: 16
- Priority: P5
- Status: Resolved
- Resolution: Fixed
- Submitted: 2020-10-13
- Updated: 2020-10-22
- Resolved: 2020-10-16
The Version table provides details related to the release that this issue/RFE will be addressed.
Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.
To download the current JDK release, click here.
Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.
To download the current JDK release, click here.
| JDK 16 |
|---|
16 b21Fixed  |
Related Reports
|
Blocks :
|
Description
Next mark bitmap clearing is supposed to abort after a concurrent mark abort. It does not because the condition is wrong.
I.e. the current code
// Abort iteration if after yielding the marking has been aborted.
if (_cm != NULL && _cm->do_yield_check() && _cm->has_aborted()) {
return true;
}
is wrong. Correct code is:
if (_cm != NULL) {
_cm->do_yield_check();
if (_cm->has_aborted) {
return true;
}
}
Currently this has no impact except for wasted time. However with JDK-8253600 it will clear the prev bitmap containing liveness information (previously passed as next bitmap into the task) as the full gc will first update the next bitmap and then swap bitmaps.
Comments
|