After upgrading from the previous 6.12.x kernel headers to linux-headers-6.12.100+deb13-common (via the trixie-security repo), the DKMS build of nvidia-current-open/550.163.01 fails during the postinst step of linux-headers-6.12.100+deb13-amd64.
The build fails with:
/var/lib/dkms/nvidia-current-open/550.163.01/build/nvidia/nv-pci.c: In function 'nv_resize_pcie_bars':
/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'
237 | r = pci_resize_resource(pci_dev, NV_GPU_BAR1, requested_size);
| ^~~~~~~~~~~~~~~~~~~
/usr/src/linux-headers-6.12.100+deb13-common/include/linux/pci.h:1421:18: note: declared here
1421 | int __must_check pci_resize_resource(struct pci_dev *dev, int i, int size,
| ^~~~~~~~~~~~~~~~~~~
pci_resize_resource() in include/linux/pci.h now takes an additional argument compared to the previous 6.12.x point release shipped in trixie.
This is an exported kernel API used by out-of-tree DKMS modules (nvidia-current-open in this case), so the change breaks module builds for anyone relying on that interface, even though it's a security update within the same stable series.
I understand DKMS module breakage is arguably a downstream/nvidia problem, but wanted to flag that this specific kernel point release changed an exported function signature that out-of-tree code depends on, in case that was unintentional for a stable-series security update, or if it should be documented/flagged for maintainers of other out-of-tree modules.
I'm attaching the full DKMS build log (/var/lib/dkms/nvidia-current-open/550.163.01/build/make.log) showing the complete failure.
System info:
$ uname -a
Linux debian 6.12.100+deb13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.12.100-1 (2026-07-30) x86_64 GNU/Linux
$ apt show libc6 | grep ^Version
Version: 2.41-12+deb13u3
$ dpkg -l | grep linux-headers-6.12.100
iF linux-headers-6.12.100+deb13-amd64 6.12.100-1 amd64 Header files for Linux 6.12.100+deb13-amd64
ii linux-headers-6.12.100+deb13-common 6.12.100-1 all Common header files for Linux 6.12.100+deb13
$ dkms status
nvidia-current-open/550.163.01, 6.12.74+deb13+1-amd64, x86_64: installed
nvidia-current-open/550.163.01, 6.12.86+deb13-amd64, x86_64: installed
nvidia-current-open/550.163.01, 6.12.88+deb13-amd64, x86_64: installed
nvidia-current-open/550.163.01, 6.12.90+deb13.1-amd64, x86_64: installed
nvidia-current-open/550.163.01, 6.12.90+deb13-amd64, x86_64: installed
nvidia-current-open/550.163.01, 6.12.94+deb13-amd64, x86_64: installed
nvidia-current-open/550.163.01, 6.12.95+deb13-amd64, x86_64: installed
nvidia-current-open/550.163.01, 6.12.96+deb13-amd64, x86_64: installed
$ dpkg -l | grep nvidia-kernel
ii nvidia-kernel-common 20240109+1 amd64 NVIDIA binary kernel module support files
Thank you for your attention to this matter.
Giuseppe