Bug 2223701

Summary:	libvma: Socketxtreme buffer reclaim leak
Product:	Red Hat Enterprise Linux 9	Reporter:	Chris Cheney <ccheney>
Component:	libvma	Assignee:	Kamal Heib <kheib>
Status:	VERIFIED ---	QA Contact:	Afom T. Michael <tmichael>
Severity:	medium	Docs Contact:
Priority:	unspecified
Version:	9.2	CC:	hwkernel-mgr, rdma-dev-team, tmichael
Target Milestone:	rc	Keywords:	Triaged
Target Release:	9.3
Hardware:	All
OS:	All
Whiteboard:
Fixed In Version:	libvma-9.8.20-2.el9	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:
Clones:	2223720 (view as bug list)		Environment:
Last Closed:		Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Chris Cheney 2023-07-18 14:23:19 UTC

libvma was recently rebased to 9.8.20 but an upstream pre-release version (9.8.31) has a memory leak fix as noted below.

Could that patch be backported to the version in RHEL?

https://github.com/Mellanox/libvma/releases

VMA 9.8.31 Pre-release

Fixed:

    Socketxtreme buffer reclaim leak

--

VMA 9.8.30 Pre-release

Added:

    Support for ConnectX-7

--

VMA 9.8.20 Latest

Fixed:

    Fix compilation issue for gcc 13.0.1

--

Comment 1 Afom T. Michael 2023-08-01 18:26:52 UTC

Marking as "Verified: Tested, SanityOnly" as testing with libvma-9.8.20-3.el9 passed.

$ grep DISTRO /etc/motd | uniq | tr -d " "
DISTRO=RHEL-9.3.0-20230801.0
$ cat /etc/redhat-release 
Red Hat Enterprise Linux release 9.3 Beta (Plow)
$ uname -r
5.14.0-347.el9.x86_64
$ rpm -q rdma-core openmpi mpitests-openmpi ucx libvma libvma-utils
rdma-core-46.0-1.el9.x86_64
openmpi-4.1.1-7.el9.x86_64
mpitests-openmpi-7.1-2.el9.x86_64
ucx-1.14.1-1.el9.1.x86_64
libvma-9.8.20-3.el9.x86_64
libvma-utils-9.8.20-3.el9.x86_64
$

vma test results on rdma-qe-40/rdma-qe-41 & Beaker job J:8140242:
5.14.0-347.el9.x86_64, rdma-core-46.0-1.el9, mlx5, ib0, ConnectX-7 & mlx5_0
    Result | Status | Test
  ---------+--------+------------------------------------
      PASS |      0 | sockperf pingpong multicast
      PASS |      0 | sockperf throughput multicast
      PASS |      0 | sockperf throughput unicast
      PASS |      0 | sockperf pingpong unicast
      PASS |      0 | sockperf (100 sockets) pingpong multicast
      PASS |      0 | sockperf (100 sockets) pingpong unicast
      PASS |      0 | sockperf pingpong multicast pkey/vlan
      PASS |      0 | sockperf pingpong unicast pkey/vlan
Checking for failures and known issues:
  no test failures

vma test results on rdma-qe-40/rdma-qe-41 & Beaker job J:8140242:
5.14.0-347.el9.x86_64, rdma-core-46.0-1.el9, mlx5, roce.45, ConnectX-7 & mlx5_1
    Result | Status | Test
  ---------+--------+------------------------------------
      PASS |      0 | sockperf pingpong multicast
      PASS |      0 | sockperf throughput multicast
      PASS |      0 | sockperf throughput unicast
      PASS |      0 | sockperf pingpong unicast
      PASS |      0 | sockperf (100 sockets) pingpong multicast
      PASS |      0 | sockperf (100 sockets) pingpong unicast
      PASS |      0 | sockperf pingpong multicast pkey/vlan
      PASS |      0 | sockperf pingpong unicast pkey/vlan

Comment 4 Afom T. Michael 2023-08-03 21:31:58 UTC

RHEL-9.3.0-20230803.31 has "libvma-9.8.20-3.el9.x86_64 & libvma-utils-9.8.20-3.el9.x86_64" and vma tests passed. Therefore, moving this to "Verified".

$ grep DISTRO /etc/motd | uniq | tr -d " "
DISTRO=RHEL-9.3.0-20230803.31
$ cat /etc/redhat-release 
Red Hat Enterprise Linux release 9.3 Beta (Plow)
$ uname -r
5.14.0-348.el9.x86_64
$ grep DISTRO /etc/motd | uniq | tr -d " "
DISTRO=RHEL-9.3.0-20230803.31
$ cat /etc/redhat-release 
Red Hat Enterprise Linux release 9.3 Beta (Plow)
$ uname -r
5.14.0-348.el9.x86_64
$ rpm -q rdma-core openmpi mpitests-openmpi ucx libvma libvma-utils
rdma-core-46.0-1.el9.x86_64
openmpi-4.1.1-7.el9.x86_64
mpitests-openmpi-7.1-2.el9.x86_64
ucx-1.14.1-1.el9.1.x86_64
libvma-9.8.20-3.el9.x86_64
libvma-utils-9.8.20-3.el9.x86_64
$ 

vma test results on rdma-dev-19/rdma-dev-20 & Beaker job J:8150465:
5.14.0-348.el9.x86_64, rdma-core-46.0-1.el9, mlx5, ib0, ConnectX-4 & mlx5_2
    Result | Status | Test
  ---------+--------+------------------------------------
      PASS |      0 | sockperf pingpong multicast
      PASS |      0 | sockperf throughput multicast
      PASS |      0 | sockperf throughput unicast
      PASS |      0 | sockperf pingpong unicast
      PASS |      0 | sockperf (100 sockets) pingpong multicast
      PASS |      0 | sockperf (100 sockets) pingpong unicast
      PASS |      0 | sockperf pingpong multicast pkey/vlan
      PASS |      0 | sockperf pingpong unicast pkey/vlan
Checking for failures and known issues:
  no test failures

vma test results on rdma-dev-19/rdma-dev-20 & Beaker job J:8150465:
5.14.0-348.el9.x86_64, rdma-core-46.0-1.el9, mlx5, roce.45, ConnectX-4 Lx & mlx5_bond_0
    Result | Status | Test
  ---------+--------+------------------------------------
      PASS |      0 | sockperf pingpong multicast
      PASS |      0 | sockperf throughput multicast
      PASS |      0 | sockperf throughput unicast
      PASS |      0 | sockperf pingpong unicast
      PASS |      0 | sockperf (100 sockets) pingpong multicast
      PASS |      0 | sockperf (100 sockets) pingpong unicast
      PASS |      0 | sockperf pingpong multicast pkey/vlan
      PASS |      0 | sockperf pingpong unicast pkey/vlan
Checking for failures and known issues:
  no test failures