JDK-7151089 : PS NUMA: NUMA allocator should not attempt to free pages when using SHM large pages
  • Type: Bug
  • Component: hotspot
  • Sub-Component: gc
  • Affected Version: hs23
  • Priority: P4
  • Status: Resolved
  • Resolution: Fixed
  • OS: linux
  • CPU: generic
  • Submitted: 2012-03-05
  • Updated: 2013-06-26
  • Resolved: 2012-03-21
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 7 JDK 8 Other
7u40Fixed 8Fixed hs24Fixed
We already turn adaptive resizing off (UseAdaptiveNUMAChunkSizing and UseAdaptiveSizePolicy) when using SHM-based large pages because these don't support uncommiting. However that is not enough - during the initialization we still do it once, which on linux leads to mmaping over the eden portion of the SHM segment. By doing this we loose large pages support and also increase memory requirements.

EVALUATION http://hg.openjdk.java.net/hsx/hotspot-main/hotspot/rev/cf956638b844

EVALUATION http://hg.openjdk.java.net/hsx/hotspot-gc/hotspot/rev/cf956638b844

SUGGESTED FIX Avoid calling os::free_memory() for SHM - based setups.

EVALUATION See description.