data built 2026-10-19T13:16:53Z

Look up another CVE · open CVE-2026-64269 in the interactive tracker

CVE-2026-64269

RDMA/rtrs-srv: Bound RDMA-Write length to chunk size in rdma_write_sg

9.1 EPSS 0.7% AV:N

Published 2026-07-25 (3 months ago) by the Linux kernel CNA. Subsystem drivers/infiniband. Debian urgency: not yet assigned.

Debian does not build this code. The per-release build detail is under "Where it lives" below.

What to do, per Debian release

Debian's own status for CVE-2026-64269 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

ReleaseStatus VersionWhat this means
Debian 12bookwormfixedDLA-4720-1 · 2026-08-05 · text6.1.180-1Update the kernel and reboot. DLA-4720-1 shipped the fix on 2026-08-05, in 6.1.180-1. A running kernel keeps the old code until the machine restarts.
Debian 13trixiefixed6.12.96-1Update the kernel and reboot. Fixed in 6.12.96-1. A running kernel keeps the old code until the machine restarts.
Debian 14forkyfixed7.1.4-1Update the kernel and reboot. Fixed in 7.1.4-1. A running kernel keeps the old code until the machine restarts.
Debian unstablesidfixed7.1.4-1Update the kernel and reboot. Fixed in 7.1.4-1. A running kernel keeps the old code until the machine restarts.
Debian 12bookworm · linux-6.12fixedDLA-4724-1 · 2026-08-07 · text6.12.100-1~deb12u1Update the kernel and reboot. DLA-4724-1 shipped the fix on 2026-08-07, in 6.12.100-1~deb12u1. A running kernel keeps the old code until the machine restarts.

The vulnerability

What the bug is

When the server answers an RTRS READ, rdma_write_sg() builds the source scatter/gather entry for the IB_WR_RDMA_WRITE that returns data to the peer. Its length is taken directly from the wire descriptor: plist->length = le32_to_cpu(id->rd_msg->desc[0].len); rd_msg points into the chunk buffer that the remote peer filled via RDMA-WRITE-WITH-IMM (rtrs_srv_rdma_done() -> process_io_req() -> process_read()), so desc[0].len is attacker-controlled and, before this change, was only rejected when zero. The source address is the fixed chunk start (dma_addr[msg_id]) and the source lkey is the PD-wide local_dma_lkey, which is not tied to the chunk's MR mapping, so the verbs layer does not constrain the transfer length to max_chunk_size. msg_id and off are bounded against queue_depth and max_chunk_size in rtrs_srv_rdma_done(), but desc[0].len is a separate field that was not checked against the chunk size. A peer that advertises desc[0].len larger than max_chunk_size can make the posted RDMA write read past the chunk's mapped region. The resulting behaviour depends on the IOMMU configuration: with no IOMMU or in passthrough mode the read may extend into memory adjacent to the chunk and be returned to the peer, which can disclose host memory; with a translating IOMMU the out-of-range access is expected to fault and abort the connection. In either case the transfer exceeds what the protocol permits and is driven by a remote peer. Reject a descriptor length above max_chunk_size, mirroring the existing off >= max_chunk_size bound in rtrs_srv_rdma_done(). Legitimate clients do not exceed it: the client sets desc[0].len to its MR length, which is capped at the negotiated max_io_size (max_chunk_size - MAX_HDR_SIZE).

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/ulp/rtrs/rtrs-srv.c

What this is
RTRS TRANSPORT DRIVERSMAINTAINERS section matched on drivers/infiniband/ulp/rtrs/ · linux-rdma@vger.kernel.org · Maintained
Kernel module
none to look for. The file would have gone into rtrs_server, and Debian does not compile it into that module, so finding rtrs_server in lsmod would say nothing about this CVE.

Check it on the machine

grep -wE 'CONFIG_INFINIBAND|CONFIG_INFINIBAND_RTRS|CONFIG_INFINIBAND_RTRS_SERVER' /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.

Show whether Debian builds this, per release

Debian does not build this code. In 5 releases below, every kernel flavour Debian publishes has the switch off, so the vulnerable file is compiled into no kernel binary Debian ships.

  • bookworm, trixie, forky, sid, bookworm · linux-6.12n
n
not enabled on any flavour - this release does not build the code
Build switches
CONFIG_INFINIBAND, CONFIG_INFINIBAND_RTRS, CONFIG_INFINIBAND_RTRS_SERVER
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
2026-07-25 (3 months ago)source
CVSS
9.1 critical
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:Hsource recompute it
Attack vector
network (AV:N)source
EPSS
0.71%, higher than 52.8% of all CVEssource
CISA KEV
not listedthe KEV catalogue
Debian urgency
not yet assignedsource
Subsystem
drivers/infiniband

Upstream fixes

The versions the CNA records as introducing and fixing this, and the commit for each. source

IntroducedFixed inCommit
5.85.10.266a35b7a8728a5
5.85.15.21268c09762172f
5.86.1.1786cada5401508
5.86.6.1452912f3d40355
5.86.12.966f40246f4312
5.86.18.395a45d0aa1fa5
5.87.1.4da3e44add94b
5.87.2963af8d97a8c

Debian advisories

AdvisoryDateShipped toDebian
DLA-4724-12026-08-07bookworm 6.12.100-1~deb12u1tracker
DLA-4723-12026-08-07bullseye 6.1.180-1~deb11u1tracker
DLA-4720-12026-08-05bookworm 6.1.180-1tracker

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:NAttack vector, network
RTRS accepts network connections over routable RDMA transports including RoCE and iWARP, and the malicious descriptor is supplied by a remote peer.
AC:LAttack complexity, low
The peer directly controls the descriptor length and destination buffer; no race, timing, or victim-controlled state is required in no-IOMMU, passthrough, or software-RDMA deployments.
PR:NPrivileges required, none
The handshake validates protocol fields but performs no credential, capability, or authentication check before publishing chunk rkeys and accepting I/O requests.
UI:NUser interaction, none
After connecting, the attacker can submit the crafted RDMA READ request without any victim action.
S:UScope, unchanged
The vulnerable RTRS module and affected host memory remain within the kernel's security authority; a translating IOMMU prevents rather than permits crossing its isolation boundary.
C:HConfidentiality impact, high
The PD-wide local DMA key does not constrain the attacker-supplied transfer length, permitting a broad out-of-bounds read of adjacent host memory that is returned to the peer.
I:NIntegrity impact, none
The malformed operation reads server memory and writes it only to the peer's registered destination; it provides no server-side memory-write primitive.
A:HAvailability impact, high
Software-RDMA can CPU-copy through invalid pages and oops or crash, while hardware or IOMMU faults can abort paths. An unauthenticated peer can trigger the condition repeatedly.