-
Bug
-
Resolution: Fixed
-
P3
-
18, 19, 20
-
b29
Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
---|---|---|---|---|---|---|
JDK-8289406 | 20 | Thomas Schatzl | P3 | Resolved | Fixed | b04 |
JDK-8291344 | 19.0.2 | Thomas Schatzl | P3 | Resolved | Fixed | b01 |
JDK-8291159 | 19.0.1 | Thomas Schatzl | P3 | Resolved | Fixed | b04 |
The BlockLocationPrinter used for decoding/printing addresses in hs_err files may call G1CollectedHeap::block_start on random addresses in the heap.
Recently, inJDK-8270540 we changed G1's BOT block_start method to reject addresses between top() and end() of a region.
This causes the hs_err file address decoder to fail, which decreases the usefulness of that hs_err file (it prints something like "error decoding...") and unnecessarily complicates backtraces (i.e. adds lots of additional stack frames to thread's backtraces when this fails).
Recently, in
This causes the hs_err file address decoder to fail, which decreases the usefulness of that hs_err file (it prints something like "error decoding...") and unnecessarily complicates backtraces (i.e. adds lots of additional stack frames to thread's backtraces when this fails).
- backported by
-
JDK-8289406 BlockLocationPrinter fails to decode addresses with G1
- Resolved
-
JDK-8291159 BlockLocationPrinter fails to decode addresses with G1
- Resolved
-
JDK-8291344 BlockLocationPrinter fails to decode addresses with G1
- Resolved
- relates to
-
JDK-8270540 G1: Refactor range checking in G1BlockOffsetTablePart::block_start* to asserts
- Resolved
(1 links to)