canonical-ubuntu-qa team mailing list archive

[Jacob Martin <2058441@xxxxxxxxxxxxxxxxxx>]

According to the document [1]: NUMA node 0 is the Grace CPU, node 1 is
the Hopper GPU, and nodes 2-8 are used for MIG [2].

[1] https://docs.nvidia.com/grace-performance-tuning-guide.pdf
[2] https://www.nvidia.com/en-us/technologies/multi-instance-gpu/

** Tags added: 6.8

** Tags added: noble

-- 
You received this bug notification because you are a member of Canonical
Platform QA Team, which is subscribed to ubuntu-kernel-tests.
https://bugs.launchpad.net/bugs/2058441

Title:
  kselftest net:reuseport_bpf_numa fails on J-nvidia-6.5 on ARM64 node
  hinyari (failed to pin to node)

Status in ubuntu-kernel-tests:
  New

Bug description:
  This issue can be found exclusively on ARM64 node "hinyari" with:
    * J-nvidia-6.5.0-1010.10
    * J-nvidia-6.5.0-1011.11 
    * J-nvidia-6.5.0-1013.13 

  The failure message is identical to bug 2029273, but I think it's
  better to file a new one here.

  Test log:
   Running 'make run_tests -C net TEST_PROGS=reuseport_bpf_numa TEST_GEN_PROGS='' TEST_CUSTOM_PROGS='''
   make: Entering directory '/home/ubuntu/autotest/client/tmp/ubuntu_kselftests_net/src/linux/tools/testing/selftests/net'
   TAP version 13
   1..1
   # timeout set to 0
   # selftests: net: reuseport_bpf_numa
   # ---- IPv4 UDP ----
   # send node 0, receive socket 0
   # ./reuseport_bpf_numa: failed to pin to node: Invalid argument
   not ok 1 selftests: net: reuseport_bpf_numa # exit=1
   make: Leaving directory '/home/ubuntu/autotest/client/tmp/ubuntu_kselftests_net/src/linux/tools/testing/selftests/net'

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu-kernel-tests/+bug/2058441/+subscriptions