Look up another CVE · open CVE-2026-68419 in the interactive tracker
CVE-2026-68419
RDMA/irdma: Prevent rereg_mr for non-mem regions
Published 2026-08-10 (2 months ago) by the Linux kernel CNA. Subsystem drivers/infiniband. Debian urgency: unimportant.
What to do, per Debian release
Debian's own status for CVE-2026-68419 in every release this site tracks, and what each one means in practice. Every status below is the one the Debian Security Team publishes for this CVE. source
| Release | Status | Version | What this means |
|---|---|---|---|
| Debian 12bookworm | not affected | 6.1.190-1 | Nothing to do. bookworm never shipped the vulnerable code. |
| Debian 13trixie | fixed | 6.12.101-1 | Update the kernel and reboot. Fixed in 6.12.101-1. A running kernel keeps the old code until the machine restarts. |
| Debian 14forky | fixed | 7.1.6-1 | Update the kernel and reboot. Fixed in 7.1.6-1. A running kernel keeps the old code until the machine restarts. |
| Debian unstablesid | fixed | 7.1.6-1 | Update the kernel and reboot. Fixed in 7.1.6-1. A running kernel keeps the old code until the machine restarts. |
| Debian 12bookworm · linux-6.12 | fixedDLA-4745-1 · 2026-08-18 · text | 6.12.101-1~deb12u1 | Update the kernel and reboot. DLA-4745-1 shipped the fix on 2026-08-18, in 6.12.101-1~deb12u1. A running kernel keeps the old code until the machine restarts. |
The vulnerability
What the bug is
When a QP/CQ/SRQ is created, a two step process is used where the buffer is allocated in userspace and explicitly registered with the normal reg_mr mechanism prior to creating the actual QP/CQ/SRQ object. These special registrations are indicated via an ABI field so the driver knows that they do not have a valid mkey and to skip the actual CQP command submission. Since these are real MR objects from the core's perspective, it is possible for a user application to invoke rereg_mr on them and cause a real CQP op to be emitted with the zero-initialized mkey value of 0. Fix this by preventing rereg_mr on these special regions.
Where it lives
The fix touches these files. If the subsystem is one you do not use, the practical exposure is lower, though the package is still the vulnerable one.
drivers/infiniband/hw/irdma/verbs.c
- What this is
- INTEL ETHERNET PROTOCOL DRIVER FOR RDMAMAINTAINERS section matched on drivers/infiniband/hw/irdma/ · linux-rdma@vger.kernel.org · Supported
- Kernel module
- irdma
Check it on the machine
grep -wE 'CONFIG_INFINIBAND|CONFIG_INFINIBAND_IRDMA' /boot/config-$(uname -r)Reads the booted kernel's own configuration, on the machine and nowhere else. =y is built in, =m is a module, and an is not set line means the code is not there.
lsmod | grep -w irdmaPrints a line if that module is loaded right now, and nothing if it is not. One that is not loaded can still be loaded later by anything that needs it.
Show whether Debian builds this, per release
Debian builds this, but not on every flavour. At least one release below builds the code on some of its kernel flavours and not on others, so the answer depends on the architecture and flavour installed.
- bookwormMnot built on amd64/cloud-amd64, arm64/cloud-arm64
- trixieMnot built on amd64/cloud-amd64, arm64/cloud-arm64, armel/rpi
- forky, sidMnot built on amd64/cloud-amd64, arm64/cloud-arm64, armhf/armmp, armhf/armmp-lpae, loong64/loong64, ppc64el/powerpc64le, ppc64el/powerpc64le-64k, riscv64/riscv64, s390x/s390x
- bookworm · linux-6.12Mnot built on amd64/cloud-amd64, arm64/cloud-arm64
- M
- a module where it is built at all; some flavours do not build it
- Build switches
- CONFIG_INFINIBAND, CONFIG_INFINIBAND_IRDMA
- Reads as
- all on means the file is compiled; any one off means it is not.
From the configuration Debian publishes for every kernel flavour it builds. It covers only the architectures Debian ships a linux-config package for, and nothing about a kernel you built yourself. Debian's own status above stays authoritative about the source package.MAINTAINERS kbuild Makefiles Debian linux-config packages
Signals
- Published
- 2026-08-10 (2 months ago)source
- CVSS
- 7.8 high
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:Hsource recompute it - Attack vector
- local (AV:L)source
- EPSS
- 0.13%, higher than 2.2% of all CVEssource
- CISA KEV
- not listedthe KEV catalogue
- Debian urgency
- unimportantsource
- Subsystem
- drivers/infiniband
Upstream fixes
The versions the CNA records as introducing and fixing this, and the commit for each. source
| Introduced | Fixed in | Commit |
|---|---|---|
| 6.6.120 | 6.6.148 | fb46d134e1b8 |
| 6.7 | 6.12.101 | b5029e91c634 |
| 6.7 | 6.18.42 | ca1c29f05274 |
| 6.7 | 7.1.6 | dbaa37e06091 |
| 6.7 | 7.2 | a846aecb931b |
Debian advisories
| Advisory | Date | Shipped to | Debian |
|---|---|---|---|
| DLA-4745-1 | 2026-08-18 | bookworm 6.12.101-1~deb12u1 | tracker |
Why the CNA scored it this way
Quoted from the kernel CNA's own scoring file, word for word. A highlighted metric carries the most severe value CVSS v3.1 defines for it, which is what pushes the score up. source
- AV:LAttack vector, local
- The vulnerability is reached through the ib_uverbs REREG_MR command on /dev/infiniband/uverbs*, a local character device; no network input reaches irdma_rereg_user_mr(). The resulting corrupt hardware STAG can later be abused over RDMA, but initial access is local.
- AC:LAttack complexity, low
- The attacker fully controls every precondition: register a buffer with reg_type IRDMA_MEMREG_TYPE_QP/CQ/SRQ, then call rereg_mr on that handle. There is no race, no timing dependency and no reliance on memory layout the attacker cannot influence.
- PR:LPrivileges required, low
- Only an open file descriptor on /dev/infiniband/uverbsN is needed; rdma-core udev rules ship these nodes world-accessible (0666) and they are routinely exposed to containers. ib_uverbs_rereg_mr() performs no capability check, only uobject ownership of the caller's own MR.
- UI:NUser interaction, none
- The attacker performs the entire sequence (reg_mr with a non-MEM reg_type, then rereg_mr) from its own process. No action by any other user or administrator is required.
- S:UScope, unchanged
- The corruption stays within the kernel/RDMA-device security authority managed by the same OS instance; there is no VM, IOMMU or hypervisor boundary crossed by the erroneous CQP command.
- C:HConfidentiality impact, high
- Programming HW STAG index 0 with attacker-chosen remote-read rights, combined with dereg_mr never invalidating it, leaves a live hardware mapping over PBLEs and umem pages that are freed and reused, giving RDMA-read access to recycled kernel/other-process memory; it can also collide with another process's MR at stag index 0.
- I:HIntegrity impact, high
- The same stale/hijacked STAG can carry IB_ACCESS_REMOTE_WRITE, and the IB_MR_REREG_TRANS path frees the PBLEs and releases the umem of a live QP/CQ/SRQ while hardware keeps DMAing into those pages, yielding device-driven writes into freed and reallocated kernel memory.
- A:HAvailability impact, high
- Issuing a CQP op with a zero mkey corrupts device state and can fail the CQP op, triggering the driver's reset path; the DMA-after-free into recycled pages and the clobbering of another region's STAG readily produce kernel memory corruption, oops or full RDMA device reset.