UBUNTU-CVE-2025-21732: Race Condition

Published Feb 27, 2025

Updated

In the Linux kernel, the following vulnerability has been resolved: RDMA/mlx5: Fix a race for an ODP MR which leads to CQE with error This patch addresses a race condition for an ODP MR that can result in a CQE with an error on the UMR QP. During the mlx5ibderegmr() flow, the following sequence of calls occurs: mlx5revokemr() mlx5rumrrevokemr() mlx5rumrpostsendwait() At this point, the lkey is freed from the hardware's perspective. However, concurrently, mlx5ibinvalidaterange() might be triggered by another task attempting to invalidate a range for the same freed lkey. This task will: - Acquire the umemodp->umemmutex lock. - Call mlx5rumrupdatexlt() on the UMR QP. - Since the lkey has already been freed, this can lead to a CQE error, causing the UMR QP to enter an error state [1]. To resolve this race condition, the umemodp->umemmutex lock is now also acquired as part of the mlx5revokemr() scope. Upon successful revoke, we set umemodp->private which points to that MR to NULL, preventing any further invalidation attempts on its lkey. [1] From dmesg: infiniband rocep8s0f0: dumpcqe:277:(pid 0): WC error: 6, Message: memory bind operation error cqedump: 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 cqedump: 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 cqedump: 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 cqedump: 00000030: 00 00 00 00 08 00 78 06 25 00 11 b9 00 0e dd d2 WARNING: CPU: 15 PID: 1506 at drivers/infiniband/hw/mlx5/umr.c:394 mlx5rumrpostsendwait+0x15a/0x2b0 [mlx5ib] Modules linked in: ip6tablemangle ip6tablenatip6tablefilter ip6tables iptablemangle xtconntrack xtMASQUERADE nfconntracknetlink nfnetlink xtaddrtype iptablenat nfnat brnetfilter rpcsecgsskrb5 authrpcgss oidregistry overlay rpcrdma rdmaucm ibiser libiscsi scsitransportiscsi rdmacm iwcm ibumad ibipoib ibcm mlx5ib ibuverbs ibcore fuse mlx5core CPU: 15 UID: 0 PID: 1506 Comm: ibvrcpingpong Not tainted 6.12.0-rc7+ #1626 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 RIP: 0010:mlx5rumrpostsendwait+0x15a/0x2b0 [mlx5ib] [..] Call Trace: <TASK> mlx5rumrupdatexlt+0x23c/0x3e0 [mlx5ib] mlx5ibinvalidaterange+0x2e1/0x330 [mlx5ib] mmunotifierinvalidaterangestart+0x1e1/0x240 zappagerangesingle+0xf1/0x1a0 madvisevmabehavior+0x677/0x6e0 domadvise+0x1a2/0x4b0 x64sysmadvise+0x25/0x30 dosyscall64+0x6b/0x140 entrySYSCALL64afterhwframe+0x76/0x7e

Affected Software

155 affected componentsFixes available

ubuntu/linux-hwe-edge*

ubuntu/linux-aws-5.0*

ubuntu/linux-aws-5.3*

ubuntu/linux-azure*

ubuntu/linux-azure-5.3*

ubuntu/linux-azure-edge*

ubuntu/linux-gcp*

ubuntu/linux-gcp-5.3*

ubuntu/linux-gke-4.15*

ubuntu/linux-gke-5.4*

ubuntu/linux-gkeop-5.4*

ubuntu/linux-hwe*

ubuntu/linux-hwe-edge*

ubuntu/linux-oem*

ubuntu/linux-oracle-5.0*

ubuntu/linux-oracle-5.3*

ubuntu/linux-aws-5.11*

ubuntu/linux-aws-5.13*

ubuntu/linux-aws-5.15*

ubuntu/linux-aws-5.8*

ubuntu/linux-azure-5.11*

ubuntu/linux-azure-5.13*

ubuntu/linux-azure-5.15*

ubuntu/linux-azure-5.8*

ubuntu/linux-azure-fde*

ubuntu/linux-azure-fde-5.15*

ubuntu/linux-gcp-5.11*

ubuntu/linux-gcp-5.13*

ubuntu/linux-gcp-5.15*

ubuntu/linux-gcp-5.8*

ubuntu/linux-gke*

ubuntu/linux-gke-5.15*

ubuntu/linux-gkeop*

ubuntu/linux-gkeop-5.15*

ubuntu/linux-hwe-5.11*

ubuntu/linux-hwe-5.13*

ubuntu/linux-hwe-5.15*

ubuntu/linux-hwe-5.8*

ubuntu/linux-ibm-5.15*

ubuntu/linux-intel-5.13*

ubuntu/linux-intel-iotg-5.15*

ubuntu/linux-lowlatency-hwe-5.15*

ubuntu/linux-nvidia-tegra-5.15*

ubuntu/linux-oem-5.10*

ubuntu/linux-oem-5.13*

ubuntu/linux-oem-5.6*

ubuntu/linux-oracle-5.11*

ubuntu/linux-oracle-5.13*

ubuntu/linux-oracle-5.15*

ubuntu/linux-oracle-5.8*

ubuntu/linux-raspi2*

ubuntu/linux-riscv*

ubuntu/linux-riscv-5.11*

ubuntu/linux-riscv-5.15*

ubuntu/linux-riscv-5.8*

ubuntu/linux*

ubuntu/linux-allwinner-5.19*

ubuntu/linux-aws*

ubuntu/linux-aws-5.19*

ubuntu/linux-aws-6.2*

ubuntu/linux-aws-6.5*

ubuntu/linux-aws-6.8<6.8.0-1032.34~22.04.1

6.8.0-1032.34~22.04.1

ubuntu/linux-azure*

ubuntu/linux-azure-5.19*

ubuntu/linux-azure-6.2*

ubuntu/linux-azure-6.5*

ubuntu/linux-azure-6.8<6.8.0-1034.39~22.04.1

6.8.0-1034.39~22.04.1

ubuntu/linux-azure-fde*

ubuntu/linux-azure-fde-5.19*

ubuntu/linux-azure-fde-6.2*

ubuntu/linux-gcp*

ubuntu/linux-gcp-5.19*

ubuntu/linux-gcp-6.2*

ubuntu/linux-gcp-6.5*

ubuntu/linux-gcp-6.8<6.8.0-1033.35~22.04.1

6.8.0-1033.35~22.04.1

ubuntu/linux-gke*

ubuntu/linux-gkeop*

ubuntu/linux-hwe-5.19*

ubuntu/linux-hwe-6.2*

ubuntu/linux-hwe-6.5*

ubuntu/linux-hwe-6.8<6.8.0-64.67~22.04.1

6.8.0-64.67~22.04.1

ubuntu/linux-ibm*

ubuntu/linux-ibm-6.8<6.8.0-1029.29~22.04.1

6.8.0-1029.29~22.04.1

ubuntu/linux-intel-iot-realtime*

ubuntu/linux-intel-iotg*

ubuntu/linux-kvm*

ubuntu/linux-lowlatency*

ubuntu/linux-lowlatency-hwe-5.19*

ubuntu/linux-lowlatency-hwe-6.2*

ubuntu/linux-lowlatency-hwe-6.5*

ubuntu/linux-lowlatency-hwe-6.8<6.8.0-64.67.1~22.04.1

6.8.0-64.67.1~22.04.1

ubuntu/linux-nvidia*

ubuntu/linux-nvidia-6.2*

ubuntu/linux-nvidia-6.5*

ubuntu/linux-nvidia-6.8<6.8.0-1031.34~22.04.1

6.8.0-1031.34~22.04.1

ubuntu/linux-nvidia-tegra*

ubuntu/linux-nvidia-tegra-igx*

ubuntu/linux-oem-5.17*

ubuntu/linux-oem-6.0*

ubuntu/linux-oem-6.1*

ubuntu/linux-oem-6.5*

ubuntu/linux-oracle*

ubuntu/linux-oracle-6.5*

ubuntu/linux-oracle-6.8<6.8.0-1029.30~22.04.1

6.8.0-1029.30~22.04.1

ubuntu/linux-raspi*

ubuntu/linux-realtime*

ubuntu/linux-riscv*

ubuntu/linux-riscv-5.19*

ubuntu/linux-riscv-6.5*

ubuntu/linux-riscv-6.8<6.8.0-64.67~22.04.1

6.8.0-64.67~22.04.1

ubuntu/linux-starfive-5.19*

ubuntu/linux-starfive-6.2*

ubuntu/linux-starfive-6.5*

ubuntu/linux-xilinx-zynqmp*

ubuntu/linux-bluefield*

ubuntu/linux-intel-iot-realtime*

ubuntu/linux-realtime*

ubuntu/linux-realtime-6.8<6.8.1-1025.26~22.04.1

6.8.1-1025.26~22.04.1

ubuntu/linux-aws-fips*

ubuntu/linux-azure-fips*

ubuntu/linux-fips*

ubuntu/linux-gcp-fips*

ubuntu/linux-aws-fips*

ubuntu/linux-azure-fips*

ubuntu/linux-fips*

ubuntu/linux-gcp-fips*

ubuntu/linux<6.8.0-64.67

6.8.0-64.67

ubuntu/linux-aws<6.8.0-1032.34

6.8.0-1032.34

ubuntu/linux-azure<6.8.0-1034.39

6.8.0-1034.39

ubuntu/linux-azure-6.11<6.11.0-1015.15~24.04.1

6.11.0-1015.15~24.04.1

ubuntu/linux-azure-nvidia<6.8.0-1022.23

6.8.0-1022.23

ubuntu/linux-gcp<6.8.0-1033.35

6.8.0-1033.35

ubuntu/linux-gcp-6.11<6.11.0-1015.15~24.04.1

6.11.0-1015.15~24.04.1

ubuntu/linux-gke<6.8.0-1028.32

6.8.0-1028.32

ubuntu/linux-gkeop<6.8.0-1015.17

6.8.0-1015.17

ubuntu/linux-hwe-6.11<6.11.0-26.26~24.04.1

6.11.0-26.26~24.04.1

ubuntu/linux-ibm<6.8.0-1029.29

6.8.0-1029.29

ubuntu/linux-lowlatency<6.8.0-64.67.1

6.8.0-64.67.1

ubuntu/linux-lowlatency-hwe-6.11<6.11.0-1014.15~24.04.1

6.11.0-1014.15~24.04.1

ubuntu/linux-nvidia<6.8.0-1031.34

6.8.0-1031.34

ubuntu/linux-nvidia-6.11<6.11.0-1010.10

6.11.0-1010.10

ubuntu/linux-nvidia-lowlatency<6.8.0-1031.34.1

6.8.0-1031.34.1

ubuntu/linux-nvidia-tegra<6.8.0-1008.8

6.8.0-1008.8

ubuntu/linux-oem-6.11<6.11.0-1022.22

6.11.0-1022.22

ubuntu/linux-oem-6.8<6.8.0-1031.31

6.8.0-1031.31

ubuntu/linux-oracle<6.8.0-1029.30

6.8.0-1029.30

ubuntu/linux-raspi<6.8.0-1031.35

6.8.0-1031.35

ubuntu/linux-raspi-realtime*

ubuntu/linux-realtime*

ubuntu/linux-riscv*

ubuntu/linux-xilinx<6.8.0-1017.18

6.8.0-1017.18

ubuntu/linux-raspi-realtime<6.8.0-2026.27

6.8.0-2026.27

ubuntu/linux-realtime<6.8.1-1025.26

6.8.1-1025.26

ubuntu/linux-fips<6.8.0-78.78+fips1

6.8.0-78.78+fips1

ubuntu/linux-vmware*

Remediation

Recommended actions to resolve this vulnerability, in priority order.

Upgrade
Upgrade ubuntu/linux-aws-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-1032.34~22.04.1
Upgrade
Upgrade ubuntu/linux-azure-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-1034.39~22.04.1
Upgrade
Upgrade ubuntu/linux-gcp-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-1033.35~22.04.1
Upgrade
Upgrade ubuntu/linux-hwe-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-64.67~22.04.1
Upgrade
Upgrade ubuntu/linux-ibm-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-1029.29~22.04.1
Upgrade
Upgrade ubuntu/linux-lowlatency-hwe-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-64.67.1~22.04.1
Upgrade
Upgrade ubuntu/linux-nvidia-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-1031.34~22.04.1
Upgrade
Upgrade ubuntu/linux-oracle-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-1029.30~22.04.1
Upgrade
Upgrade ubuntu/linux-riscv-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-64.67~22.04.1
Upgrade
Upgrade ubuntu/linux-realtime-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.1-1025.26~22.04.1
Upgrade
Upgrade ubuntu/linux to a version that resolves this vulnerability.
Fixed in 6.8.0-64.67
Upgrade
Upgrade ubuntu/linux-aws to a version that resolves this vulnerability.
Fixed in 6.8.0-1032.34
Upgrade
Upgrade ubuntu/linux-azure to a version that resolves this vulnerability.
Fixed in 6.8.0-1034.39
Upgrade
Upgrade ubuntu/linux-azure-6.11 to a version that resolves this vulnerability.
Fixed in 6.11.0-1015.15~24.04.1
Upgrade
Upgrade ubuntu/linux-azure-nvidia to a version that resolves this vulnerability.
Fixed in 6.8.0-1022.23
Upgrade
Upgrade ubuntu/linux-gcp to a version that resolves this vulnerability.
Fixed in 6.8.0-1033.35
Upgrade
Upgrade ubuntu/linux-gcp-6.11 to a version that resolves this vulnerability.
Fixed in 6.11.0-1015.15~24.04.1
Upgrade
Upgrade ubuntu/linux-gke to a version that resolves this vulnerability.
Fixed in 6.8.0-1028.32
Upgrade
Upgrade ubuntu/linux-gkeop to a version that resolves this vulnerability.
Fixed in 6.8.0-1015.17
Upgrade
Upgrade ubuntu/linux-hwe-6.11 to a version that resolves this vulnerability.
Fixed in 6.11.0-26.26~24.04.1
Upgrade
Upgrade ubuntu/linux-ibm to a version that resolves this vulnerability.
Fixed in 6.8.0-1029.29
Upgrade
Upgrade ubuntu/linux-lowlatency to a version that resolves this vulnerability.
Fixed in 6.8.0-64.67.1
Upgrade
Upgrade ubuntu/linux-lowlatency-hwe-6.11 to a version that resolves this vulnerability.
Fixed in 6.11.0-1014.15~24.04.1
Upgrade
Upgrade ubuntu/linux-nvidia to a version that resolves this vulnerability.
Fixed in 6.8.0-1031.34
Upgrade
Upgrade ubuntu/linux-nvidia-6.11 to a version that resolves this vulnerability.
Fixed in 6.11.0-1010.10
Upgrade
Upgrade ubuntu/linux-nvidia-lowlatency to a version that resolves this vulnerability.
Fixed in 6.8.0-1031.34.1
Upgrade
Upgrade ubuntu/linux-nvidia-tegra to a version that resolves this vulnerability.
Fixed in 6.8.0-1008.8
Upgrade
Upgrade ubuntu/linux-oem-6.11 to a version that resolves this vulnerability.
Fixed in 6.11.0-1022.22
Upgrade
Upgrade ubuntu/linux-oem-6.8 to a version that resolves this vulnerability.
Fixed in 6.8.0-1031.31
Upgrade
Upgrade ubuntu/linux-oracle to a version that resolves this vulnerability.
Fixed in 6.8.0-1029.30
Upgrade
Upgrade ubuntu/linux-raspi to a version that resolves this vulnerability.
Fixed in 6.8.0-1031.35
Upgrade
Upgrade ubuntu/linux-xilinx to a version that resolves this vulnerability.
Fixed in 6.8.0-1017.18
Upgrade
Upgrade ubuntu/linux-raspi-realtime to a version that resolves this vulnerability.
Fixed in 6.8.0-2026.27
Upgrade
Upgrade ubuntu/linux-realtime to a version that resolves this vulnerability.
Fixed in 6.8.1-1025.26
Upgrade
Upgrade ubuntu/linux-fips to a version that resolves this vulnerability.
Fixed in 6.8.0-78.78+fips1
Compensating control
In the mlx5 ODP MR revoke path, acquire the umem_odp->umem_mutex lock within mlx5_revoke_mr(); after a successful revoke, set umem_odp->private (the pointer to the MR) to NULL to prevent further invalidation attempts on the freed lkey.

Event History

Feb 27, 2025

CVE Published

via OSV·03:15 AM

Data Sourced

via OSV·03:15 AM

DescriptionSeverityAffected Software

Frequently Asked Questions

Which systems are realistically exposed to this issue?

Systems using the affected Ubuntu kernel packages and RDMA with Mellanox mlx5 hardware are the relevant population. The race specifically involves on-demand paging (ODP) memory regions and the mlx5 RDMA driver's UMR queue pair.

What conditions are needed to trigger the failure?

A memory region deregistration must race with another task invalidating a range for the same ODP memory region. The invalidation can attempt to use an lkey after it has been freed by the hardware, causing an error completion on the UMR queue pair.

How can I tell whether this has occurred?

The kernel log may contain mlx5 RDMA completion errors, including messages such as "WC error: 6" and "Message: memory bind operation error." The failure can place the UMR queue pair into an error state.

What does the fix change?

The fix holds the ODP memory-region mutex during the revoke operation and clears the ODP private pointer after a successful revoke. This prevents later invalidation attempts from operating on the freed lkey.

UBUNTU-CVE-2025-21732: Race Condition

Affected Software

Remediation

Event History

Frequently Asked Questions

Which systems are realistically exposed to this issue?

What conditions are needed to trigger the failure?

How can I tell whether this has occurred?

What does the fix change?

Company

Resources

Contact

Frequently Asked Questions

Which systems are realistically exposed to this issue?

What conditions are needed to trigger the failure?

How can I tell whether this has occurred?

What does the fix change?