canonical-ubuntu-qa team mailing list archive

[Koichiro Den <2089768@xxxxxxxxxxxxxxxxxx>]

Patch sent to ML: https://lists.ubuntu.com/archives/kernel-
team/2024-December/155919.html

SRU Justification:

[Impact]

Since upstream commit 003af997c8a9 ("hugetlb: force allocating surplus
hugepages on mempolicy allowed nodes"), the LTP hugetlb test case
'hugemmap10' has started failing frequently. In some testing environments,
it fails almost consistently. This occurs because the surplus allocation
behavior was changed, typically leading to allocations being made on the
least numbered possible NUMA node, even when allocation from the NUMA node
on which the current task running is feasible. This not only causes LTP
test failures but also indicates potential performance degradation.

This patch restores the previous behavior while preserving the benefits
introduced by commit 003af997c8a9. It not only reduces LTP test failure
noise but also prevents performance degradations.

[Fix]

The following upstream commit resolves the issue:
https://lore.kernel.org/all/20241204165503.628784-1-koichiro.den@xxxxxxxxxxxxx
It does not currently meet stable criteria, so it's unlikely that it will
land in upstream stable tree. That is why this SRU patch.

[Test Plan]

Run the LTP hugemmap10 and verify that it passes consistently on machines
where it previously failed. If such machines are not available, compare
behavior with and without the patch by performing the following steps:
1. prepare a machine which has two or more NUMA nodes, each having usable
   memory
2. run hugemmap10 with naively pinning it onto some CPU of NUMA node #1;
   $ sudo taskset -c {some-cpu-on-node-1} ./hugemmap10
Without this patch, you will observe consistent failures, while with this
patch, the issue will disappear.

[Where problems could occur]

This change affects only hugetlb. If any regressions are found, they would
likely impact hugetlb users.

-- 
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/2089768

Title:
  hugemmap10 from ubuntu_ltp_stable.hugetlb failed on Noble

Status in ubuntu-kernel-tests:
  New

Bug description:
  Test hugemmap10 of ubuntu_ltp_stable.hugetlb was observed to fail:
    * N-lowlatency-6.8.0-50.51.1 (drapion, amd64)
    * N-nvidia-lowlatency-6.8.0-1019.21.1 (bunsen, amd64)
    * J-linux-lowlatency-hwe-6.8-6.8.0-50.51.1~22.04.1 (recht, arm64)

  ----
  07:35:55 DEBUG| [stdout] startup='Tue Nov 26 07:35:38 2024'
  07:35:55 DEBUG| [stdout] tst_hugepage.c:84: TINFO: 3 hugepage(s) reserved
  07:35:55 DEBUG| [stdout] tst_tmpdir.c:316: TINFO: Using /tmp/ltp-8CAxsmEARh/LTP_hugLIyG9r as tmpdir (ext2/ext3/ext4 filesystem)
  07:35:55 DEBUG| [stdout] tst_test.c:1085: TINFO: Mounting none to /tmp/ltp-8CAxsmEARh/LTP_hugLIyG9r/hugetlbfs fstyp=hugetlbfs flags=0
  07:35:55 DEBUG| [stdout] tst_test.c:1860: TINFO: LTP version: 20230929-874-gba610da01
  07:35:55 DEBUG| [stdout] tst_test.c:1864: TINFO: Tested kernel: 6.8.0-50-lowlatency #51.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Nov 21 12:44:17 UTC 2024 x86_64
  07:35:55 DEBUG| [stdout] tst_test.c:1703: TINFO: Timeout per run is 0h 00m 30s
  07:35:55 DEBUG| [stdout] hugemmap10.c:388: TINFO: Base pool size: 0
  07:35:55 DEBUG| [stdout] hugemmap10.c:315: TINFO: Clean...
  07:35:55 DEBUG| [stdout] hugemmap10.c:366: TINFO: OK
  07:35:55 DEBUG| [stdout] hugemmap10.c:315: TINFO: Untouched, shared...
  07:35:55 DEBUG| [stdout] hugemmap10.c:366: TINFO: OK
  07:35:55 DEBUG| [stdout] hugemmap10.c:315: TINFO: Untouched, private...
  07:35:55 DEBUG| [stdout] hugemmap10.c:366: TINFO: OK
  07:35:55 DEBUG| [stdout] hugemmap10.c:315: TINFO: Touched, shared...
  07:35:55 DEBUG| [stdout] hugemmap10.c:366: TINFO: OK
  07:35:55 DEBUG| [stdout] hugemmap10.c:315: TINFO: Touched, private...
  07:35:55 DEBUG| [stdout] hugemmap10.c:366: TINFO: OK
  07:35:55 DEBUG| [stdout] hugemmap10.c:388: TINFO: Base pool size: 1
  07:35:55 DEBUG| [stdout] hugemmap10.c:315: TINFO: Clean...
  07:35:55 DEBUG| [stdout] hugemmap10.c:366: TINFO: OK
  07:35:55 DEBUG| [stdout] hugemmap10.c:315: TINFO: Untouched, shared...
  07:35:55 DEBUG| [stdout] hugemmap10.c:333: TFAIL: While doing munmap shared after touch: Bad HugePages_Total: expected 1, actual 2
  07:35:55 DEBUG| [stdout] hugemmap10.c:333: TFAIL: While doing munmap shared after touch: Bad HugePages_Free: expected 1, actual 2
  07:35:55 DEBUG| [stdout] hugemmap10.c:333: TFAIL: While doing munmap shared after touch: Bad HugePages_Surp: expected 0, actual 1
  07:35:55 DEBUG| [stdout]
  07:35:55 DEBUG| [stdout] Summary:
  07:35:55 DEBUG| [stdout] passed   0
  07:35:55 DEBUG| [stdout] failed   3
  07:35:55 DEBUG| [stdout] broken   0
  07:35:55 DEBUG| [stdout] skipped  0
  07:35:55 DEBUG| [stdout] warnings 0
  07:35:55 DEBUG| [stdout] tag=hugemmap10 stime=1732606538 dur=0 exit=exited stat=1 core=no cu=0 cs=4

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu-kernel-tests/+bug/2089768/+subscriptions