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

[bugproxy]

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 <email address hidden> - 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 <email address hidden> - 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.

[bugproxy (bugproxy)]

------- Comment From <email address hidden> 2017-07-18 16:45 EDT-------
Canonical, any update on this? Thank you.

[Tiago Stürmer Daitx (tdaitx)]

Sorry for the lateness, for some reason the bugproxy emails from different issues are being grouped together by gmail, which caused it to be incorrectly labeled by my filters - should be fixed now.

This has been added to the OpenJDK 8 update that is under review by the security team. It will be included in the next OpenJDK 8 upload, which should happen somewhere next week.

[bugproxy (bugproxy)]

------- Comment From <email address hidden> 2017-07-20 17:55 EDT-------
Thanks, Daitx

[Launchpad Janitor (janitor)]

This bug was fixed in the package openjdk-8 - 8u141-b15-1

---------------
openjdk-8 (8u141-b15-1) unstable; urgency=high

  * Update to 8u141-b15, Hotspot 8u141-b16 for AArch64.
  * Security fixes from 8u141:
    - CVE-2017-10102, S8163958: Improved garbage collection.
    - CVE-2017-10053, S8169209: Improved image post-processing steps.
    - CVE-2017-10067, S8169392: Additional jar validation steps.
    - CVE-2017-10081, S8170966: Right parenthesis issue.
    - CVE-2017-10078, S8171539: Better script accessibility for JavaScript.
    - CVE-2017-10087, S8172204: Better Thread Pool execution.
    - CVE-2017-10089, S8172461: Service Registration Lifecycle.
    - CVE-2017-10090, S8172465: Better handling of channel groups.
    - CVE-2017-10096, S8172469: Transform Transformer Exceptions.
    - CVE-2017-10101, S8173286: Better reading of text catalogs.
    - CVE-2017-10107, S8173697: Less Active Activations.
    - CVE-2017-10074, S8173770: Image conversion improvements.
    - CVE-2017-10110, S8174098: Better image fetching.
    - CVE-2017-10108, S8174105: Better naming attribution.
    - CVE-2017-10109, S8174113: Better sourcing of code.
    - CVE-2017-10115, S8175106: Higher quality DSA operations.
    - CVE-2017-10118, S8175110: Higher quality ECDSA operations.
    - CVE-2017-10116, S8176067: Proper directory lookup processing.
    - CVE-2017-10135, S8176760: Better handling of PKCS8 material.
    - CVE-2017-10176, S8178135: Additional elliptic curve support.
    - CVE-2017-10193, S8179101: Improve algorithm constraints implementation.
    - CVE-2017-10198, S8179998: Clear certificate chain connections.
    - S8174770: Check registry registration location.
    - S8174873: Improved certificate procesing.
    - S8176055: JMX diagnostic improvements.
    - S8176536: Improved algorithm constraints checking.
    - S8181420: PPC: Image conversion improvements.
    - S8182054: Improve wsdl support.
    - S8184185: Rearrange MethodHandle arrangements.

  [ Matthias Klose ]
  * Provide jvmdir symlink in /usr/lib/debug. Closes: #867314.
  * Fix pt_BR translation in awt message. Closes: #863331.

  [ Tiago Stürmer Daitx ]
  * debian/rules:
    - enable apport hook on Ubuntu and derivatives only.
    - remove with_zenhai logic.
    - remove unused with_tzdata logic, move tzdata build dependency
      to control.in.
    - add Breaks:tzdata-java except for wheezy, jessie or trusty.
    - re-enable jamvm for Xenial only.
    - run debian/control before build so we won't build with a invalid
      control file.
    - remove logic to select between ttf or font packages and depend
      on fonts-wqy-microhei and fonts-wqy-zenhei instead
  * debian/apport-hook.py: add an apport hook to include conffiles
    modified by the user on any report and the hs_err log file on
    crash report only. LP: #1696886.
  * patches/fontconfig-arphic-uming.diff: only enabled when with_zenhai
    was false; not required since lenny.
  * patches/hotspot-ppc64el-S8175813-mbind-invalid-argument.patch: prevent
    invalid argument message when invoking UseNUMA on a system with
    non-consecutive numa topology. LP: #1697348.

 -- Matthias Klose <email address hidden> Fri, 21 Jul 2017 1...