openjdk team mailing list archive

[bugproxy <bugproxy@xxxxxxxxxx>]

Public bug reported:

Problem Description
========================
Java throws "mbind: Invalid argument" message when invoked with UseNUMA flag in certain numa topologies with non-consecutive nodes.

 
---uname output---
not relevant
 
---Additional Hardware Info---
The JVM code used to discover which numa nodes it can bind assumes that nodes are consecutive and tries to bind from 0 to numa_max_node(), i.e. 0 to the highest node number available on the system. However, at least on PPC64 that assumption is not always true. For instance, the following numa 
topology: 

available: 4 nodes (0-1,16-17) 
node 0 cpus: 0 8 16 24 32 
node 0 size: 130706 MB 
node 0 free: 145 MB 
node 1 cpus: 40 48 56 64 72 
node 1 size: 0 MB 
node 1 free: 0 MB 
node 16 cpus: 80 88 96 104 112 
node 16 size: 130630 MB 
node 16 free: 529 MB 
node 17 cpus: 120 128 136 144 152 
node 17 size: 0 MB 
node 17 free: 0 MB 
node distances: 
node 0 1 16 17 
0: 10 20 40 40 
1: 20 10 40 40 
16: 40 40 10 20 
17: 40 40 20 10  

 
Machine Type = VM and BM 

Steps to Reproduce
=============================
 On an topology similar to the aforementioned numa topology just issue:
"java -XX:+UseParallelGC -XX:+UseNUMA -version"
 
== Comment: #1 - Gustavo Bueno Romero <gromero@xxxxxxxxxx> - 2017-06-09 10:19:37 ==
Bug is already fixed upstream:

https://bugs.openjdk.java.net/browse/JDK-8181462
http://hg.openjdk.java.net/jdk8u/jdk8u-dev/hotspot/rev/4a575a49e938

== Comment: #3 - Gustavo Bueno Romero <gromero@xxxxxxxxxx> - 2017-06-09 10:22:03 ==
Dear Daitx, could you please apply that 8u upstream fix on the next CPU cycle (around 18 July), please? Thank you. Regards.

** Affects: openjdk-8 (Ubuntu)
     Importance: Undecided
     Assignee: Ubuntu on IBM Power Systems Bug Triage (ubuntu-power-triage)
         Status: New


** Tags: architecture-ppc64le bugnameltc-155628 severity-medium targetmilestone-inin---

** Tags added: architecture-ppc64le bugnameltc-155628 severity-medium
targetmilestone-inin---

** Changed in: ubuntu
     Assignee: (unassigned) => Ubuntu on IBM Power Systems Bug Triage (ubuntu-power-triage)

** Package changed: ubuntu => openjdk-8 (Ubuntu)

-- 
You received this bug notification because you are a member of OpenJDK,
which is subscribed to openjdk-8 in Ubuntu.
https://bugs.launchpad.net/bugs/1697348

Title:
  Java throws "mbind: Invalid argument" message when invoked with
  UseNUMA

Status in openjdk-8 package in Ubuntu:
  New

Bug description:
  Problem Description
  ========================
  Java throws "mbind: Invalid argument" message when invoked with UseNUMA flag in certain numa topologies with non-consecutive nodes.

   
  ---uname output---
  not relevant
   
  ---Additional Hardware Info---
  The JVM code used to discover which numa nodes it can bind assumes that nodes are consecutive and tries to bind from 0 to numa_max_node(), i.e. 0 to the highest node number available on the system. However, at least on PPC64 that assumption is not always true. For instance, the following numa 
  topology: 

  available: 4 nodes (0-1,16-17) 
  node 0 cpus: 0 8 16 24 32 
  node 0 size: 130706 MB 
  node 0 free: 145 MB 
  node 1 cpus: 40 48 56 64 72 
  node 1 size: 0 MB 
  node 1 free: 0 MB 
  node 16 cpus: 80 88 96 104 112 
  node 16 size: 130630 MB 
  node 16 free: 529 MB 
  node 17 cpus: 120 128 136 144 152 
  node 17 size: 0 MB 
  node 17 free: 0 MB 
  node distances: 
  node 0 1 16 17 
  0: 10 20 40 40 
  1: 20 10 40 40 
  16: 40 40 10 20 
  17: 40 40 20 10  

   
  Machine Type = VM and BM 

  Steps to Reproduce
  =============================
   On an topology similar to the aforementioned numa topology just issue:
  "java -XX:+UseParallelGC -XX:+UseNUMA -version"
   
  == Comment: #1 - Gustavo Bueno Romero <gromero@xxxxxxxxxx> - 2017-06-09 10:19:37 ==
  Bug is already fixed upstream:

  https://bugs.openjdk.java.net/browse/JDK-8181462
  http://hg.openjdk.java.net/jdk8u/jdk8u-dev/hotspot/rev/4a575a49e938

  == Comment: #3 - Gustavo Bueno Romero <gromero@xxxxxxxxxx> - 2017-06-09 10:22:03 ==
  Dear Daitx, could you please apply that 8u upstream fix on the next CPU cycle (around 18 July), please? Thank you. Regards.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/openjdk-8/+bug/1697348/+subscriptions