Version: 550.163.01-2

Severity: important

Subject:nvidia-open-kernel-dkms: fails to build against
linux-image-6.12.100+deb13-amd64 (pci_resize_resource API change)

After installing the security update to Linux kernel 6.12.100-1 on Debian
13 (Trixie), DKMS fails to build the NVIDIA Open Kernel Module.

The installation of the kernel package stops with:

Error! Bad return status for module build on kernel: 6.12.100+deb13-amd64
(x86_64)

dpkg: error processing package linux-image-6.12.100+deb13-amd64
(--configure):
grep -n "error:" /var/lib/dkms/nvidia-current-open/550.163.01/build/make.log
942:/var/lib/dkms/nvidia-current-open/550.163.01/build/nvidia/nv-pci.c:237:9:
error: too few arguments to function ‘pci_resize_resource’


The relevant compiler error is:

/var/lib/dkms/nvidia-current-open/550.163.01/build/nvidia/nv-pci.c:237:9:
error: too few arguments to function 'pci_resize_resource'

As a consequence:

    linux-image-6.12.100+deb13-amd64 remains unconfigured;
    linux-image-amd64 cannot be configured;
    linux-headers-6.12.100+deb13-amd64 cannot be configured;
    the package upgrade ends with a dpkg error.

The issue is reproducible on multiple independent Debian 13 systems with
the same package versions.

Working configuration:

    linux-image-6.12.96+deb13-amd64
    nvidia-open-kernel-dkms 550.163.01-2

Failing configuration:

    linux-image-6.12.100+deb13-amd64
    nvidia-open-kernel-dkms 550.163.01-2

Steps to reproduce:

    Install Debian 13 (Trixie) with nvidia-open-kernel-dkms 550.163.01-2.
    Update to linux-image-6.12.100-1 from trixie-security.
    During DKMS autoinstall the module build fails with the error above.

Please let me know if additional logs or testing are required.

The complete DKMS build log is attached.

I can confirm this bug on independent hardware, and I have some additionalanalysis suggesting it is systematic rather than configuration-specific.

Reproduction
------------
Same failure, different machine:

  nvidia-open-kernel-dkms  550.163.01-2
  dkms                     3.2.2-1~deb13u1
  gcc                      Debian 14.2.0-19 (14.2.0)
  Debian GNU/Linux 13 (trixie)
  GPU: NVIDIA GA106M [GeForce RTX 3060 Mobile / Max-Q] [10de:2520]
       (hybrid/Optimus alongside Intel Alder Lake-P Iris Xe)

Build against 6.12.100+deb13-amd64 fails at the same location:

  nvidia/nv-pci.c:237:9: error: too few arguments to function
  'pci_resize_resource'

Building the same source against 6.12.96+deb13-amd64 succeeds.

Root cause
----------
The prototype gained a fourth parameter between the two kernels.

6.12.96, /usr/src/linux-headers-6.12.96+deb13-common/include/linux/pci.h:1421

  int __must_check pci_resize_resource(struct pci_dev *dev, int i, int size);

6.12.100, /usr/src/linux-headers-6.12.100+deb13-common/include/linux/pci.h:1421

  int __must_check pci_resize_resource(struct pci_dev *dev, int i, int size,
                                       int exclude_bars);

Why this is not configuration-dependent
---------------------------------------
This is the part I think is worth adding to the report: the driver has no
compatibility shim for this symbol at all.

1. pci_resize_resource appears exactly once in the unpacked source tree:

     $ grep -rn pci_resize_resource . --include='*.c' --include='*.h'
     ./nvidia/nv-pci.c:237:    r = pci_resize_resource(pci_dev, NV_GPU_BAR1, requested_size);

2. conftest.sh does not probe it:

     $ grep -c pci_resize_resource conftest.sh
     0

   ...while it does probe many other PCI symbols (pci_bus_address,
   pci_channel_state, pci_enable_msix_range, pci_get_domain_bus_and_slot,
   pci_driver_has_driver_managed_dma, and others), which is the mechanism
   the driver normally uses to absorb this kind of API change.

3. The call site is not inside any preprocessor conditional. The nearest
   #endif above it closes an unrelated
   NV_IS_EXPORT_SYMBOL_PRESENT_pci_find_host_bridge block.

So the call is unconditional and there is no code path that can adapt to the
new signature. Any build of 550.163.01 against a kernel carrying the
4-argument prototype will fail identically, regardless of GPU model, Optimus
versus discrete-only, or local configuration.

Scope across suites
-------------------
No currently packaged revision fixes this:

  bookworm          535.261.03-1
  trixie            550.163.01-2
  trixie-backports  550.163.01-4~bpo13+1
  sid               550.163.01-4
  experimental      555.58.02-2

trixie-backports is worth calling out explicitly, since it is the natural
first thing an affected user will reach for. It is the same upstream
550.163.01, and the Debian revisions since -2 do not touch this code path:
per the changelog, -3 backported the NV_DMA_BUF_HAS_DYNAMIC_ATTACHMENT
changes to fix the open kernel module build for Linux 6.16, and -4 was a
sync with src:nvidia-graphics-drivers. Installing the backport rebuilds the
same source and fails the same way.

Impact
------
The DKMS failure makes the linux-image postinst exit non-zero, which leaves
the kernel packages unconfigured and blocks dpkg:

  iF  linux-image-6.12.100+deb13-amd64
  iU  linux-image-amd64
  iU  linux-headers-6.12.100+deb13-amd64
  iU  linux-headers-amd64

Users who reboot into 6.12.100 before noticing will come up with no NVIDIA
driver. On this machine the other DKMS modules (acpi-call, kbdlight, xtrx)
all built successfully against 6.12.100; nvidia-current-open was the only
failure.

I have not checked whether nvidia-kernel-dkms (non-free) shares this call
site and is affected in the same way. It is not installed here, so I cannot
confirm either way.

Workaround
----------
Staying on 6.12.96+deb13-amd64, with linux-image-amd64 and
linux-headers-amd64 held, until a fixed driver is available.

*** Reporter, please consider answering these questions, where appropriate ***

   * What led up to the situation?
sudo apt upgrade
install linux-image-6.12.100+deb13-amd64
that trigger build of module in nvidia-open-kernel-dkms
   * What exactly did you do (or not do) that was effective (or
I retested with 550.163.01-4~bpo13+1 (stable-backports)
same results with k6.12.100
     ineffective)?
the module build like with K6.12.63+deb13-amd64 and 6.12.90+deb13.1-amd64
On next post I will add the make.log
_resize_resource'
   * What outcome did you expect instead?

*** End of the template - remove these template lines ***

happen also with kernel 6.12.101+deb13-amd64
same error

tags 1143204 + patch

I submitted https://salsa.debian.org/nvidia-team/nvidia-open-gpu-kernel-modules/-/merge_requests/3
as the patch to address this issue.

Hello, I uploaded the NMU to DELAYED/7 under the following name:

nvidia-open-gpu-kernel-modules_550.163.01-4.1_source.changes

While we really need this fix in trixie-updates, it needs to arrive in
unstable first per Debian policy. After that I can propose a stable
update to fix this issue in trixie to have it work with the newer
kernels in trixie-security.

Thanks,

Agathe.

We believe that the bug you reported is fixed in the latest version of
nvidia-open-gpu-kernel-modules, which is due to be installed in the Debian FTP archive.

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to 1143204@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Agathe Porte <gagath@debian.org> (supplier of updated nvidia-open-gpu-kernel-modules package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@ftp-master.debian.org)
Format: 1.8
Date: Sun, 20 Sep 2026 08:57:48 +0200
Source: nvidia-open-gpu-kernel-modules
Architecture: source
Version: 550.163.01-4.1
Distribution: unstable
Urgency: medium
Maintainer: Debian NVIDIA Maintainers <pkg-nvidia-devel@lists.alioth.debian.org>
Changed-By: Agathe Porte <gagath@debian.org>
Closes: 1143204
Changes:
 nvidia-open-gpu-kernel-modules (550.163.01-4.1) unstable; urgency=medium
 .
   [ Agathe Porte ]
   * Non-maintainer upload.
 .
   [ Ciel Avenir ]
   * Backport NV_PCI_RESIZE_RESOURCE_HAS_EXCLUDE_BARS_ARG changes from
     590.48.01 to fix open kernel module build for Linux 6.12.100.
     (Closes: #1143204)
Checksums-Sha1:
 95a98ad992c7a8a18469ae9d3d545fe67a68713c 2689 nvidia-open-gpu-kernel-modules_550.163.01-4.1.dsc
 e24acf3073dc7cf8704422d1a8d5d2d34ac1e621 37792 nvidia-open-gpu-kernel-modules_550.163.01-4.1.debian.tar.xz
 b9bab020fe53128d1ef3f58adece16749ecc7321 5736 nvidia-open-gpu-kernel-modules_550.163.01-4.1_source.buildinfo
Checksums-Sha256:
 244a0467c6baa625bc7810ec0d90aec9e2e94c6f1db19da9e94c199fa3c7d43b 2689 nvidia-open-gpu-kernel-modules_550.163.01-4.1.dsc
 2a8c66934e00e901e7ea9fb63eebf9b82beb7c0fcb3cfc48523fc3521aaf6b07 37792 nvidia-open-gpu-kernel-modules_550.163.01-4.1.debian.tar.xz
 9442ee224212125ccd3d27158d1eabdf96ce3c2fd09a9ec5c4cecc8bf8a33340 5736 nvidia-open-gpu-kernel-modules_550.163.01-4.1_source.buildinfo
Files:
 764e512235c3045f0c846a36de020d96 2689 contrib/kernel optional nvidia-open-gpu-kernel-modules_550.163.01-4.1.dsc
 66ef4f85a8b2a1205015be1e2bdbf5bf 37792 contrib/kernel optional nvidia-open-gpu-kernel-modules_550.163.01-4.1.debian.tar.xz
 fa28f43f064158454d659e162dd56f32 5736 contrib/kernel optional nvidia-open-gpu-kernel-modules_550.163.01-4.1_source.buildinfo
-----BEGIN PGP SIGNATURE-----
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=rbYM
-----END PGP SIGNATURE-----