JDK-6740923 : NUMA allocator: Ensure the progress of adaptive chunk resizing
  • Type: Bug
  • Component: hotspot
  • Sub-Component: gc
  • Affected Version: hs14
  • Priority: P4
  • Status: Resolved
  • Resolution: Fixed
  • OS: generic
  • CPU: generic
  • Submitted: 2008-08-25
  • Updated: 2010-04-03
  • Resolved: 2008-10-07
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.
JDK 6 JDK 7 Other
6u14Fixed 7Fixed hs14Fixed
The NUMA allocator resizes its lgroup chunks proportionally to the allocation rate in each chunk. However, in case when allocation fails the hole which it left is still considered to be a free space and is not taken into account when computing the allocation rate. Considering that the lgroup chunk can get as small as page it is possible that the allocation will fail immediately thus producting the allocation rate value of zero. This in turn would prevent the chunk from growing.

EVALUATION http://hg.openjdk.java.net/jdk7/hotspot-gc/hotspot/rev/06df86c2ec37

EVALUATION We should account for failed allocation requests. If the allocation requestion fails in an lgroup chunk it should be considered fully used.