I ran an update on my host today and it included an update from kernel 5.0 to 5.3.
I was using the nvidia-driver package to allow the use of my NVIDIA cards in a BOINC container.
I get this upon running: apt install nvidia-driver
I am getting this error in the build log /var/lib/dkms/nvidia-current/418.74/build/make.log
I found this which seems to indicate this might be a kernel build issue:
https://github.com/clearlinux/distribution/issues/1306
Specifically this: https://github.com/clearlinux/distribution/issues/1306#issuecomment-547306578
I was using the nvidia-driver package to allow the use of my NVIDIA cards in a BOINC container.
I get this upon running: apt install nvidia-driver
Code:
Setting up nvidia-kernel-dkms (418.74-1) ...
Loading new nvidia-current-418.74 DKMS files...
Building for 5.3.13-1-pve
Building initial module for 5.3.13-1-pve
Error! Bad return status for module build on kernel: 5.3.13-1-pve (x86_64)
Consult /var/lib/dkms/nvidia-current/418.74/build/make.log for more information.
dpkg: error processing package nvidia-kernel-dkms (--configure):
installed nvidia-kernel-dkms package post-installation script subprocess returned error exit status 10I am getting this error in the build log /var/lib/dkms/nvidia-current/418.74/build/make.log
Code:
DKMS make.log for nvidia-current-418.74 for kernel 5.3.13-1-pve (x86_64)
Sat 14 Dec 2019 12:25:53 PM CST
make V=1 -C /lib/modules/5.3.13-1-pve/build M=/var/lib/dkms/nvidia-current/418.74/build ARCH=x86_64 NV_KERNEL_SOURCES=/lib/modules/5.3.13-1-pve/build NV_KERNEL_OUTPUT=/lib/modules/5.3.13-1-pve/build NV_KER$
...
In file included from /var/lib/dkms/nvidia-current/418.74/build/nvidia/nv_uvm_interface.c:21:
/var/lib/dkms/nvidia-current/418.74/build/nvidia/nv_uvm_interface.c: In function ‘nvUvmInterfaceDeRegisterUvmOps’:
/var/lib/dkms/nvidia-current/418.74/build/common/inc/nv-linux.h:733:21: error: void value not ignored as it ought to be
int __ret = on_each_cpu(func, info, 1); \
^~~~~~~~~~~
/var/lib/dkms/nvidia-current/418.74/build/nvidia/nv_uvm_interface.c:991:5: note: in expansion of macro ‘NV_ON_EACH_CPU’
NV_ON_EACH_CPU(flush_top_half, NULL);
^~~~~~~~~~~~~~
gcc -Wp,-MD,/var/lib/dkms/nvidia-current/418.74/build/nvidia-uvm/.uvm8_va_space.o.d -nostdinc -isystem /usr/lib/gcc/x86_64-linux-gnu/8/include -I./arch/x86/include -I./arch/x86/include/generated -I./in$
gcc -Wp,-MD,/var/lib/dkms/nvidia-current/418.74/build/nvidia-uvm/.uvm8_va_space_mm.o.d -nostdinc -isystem /usr/lib/gcc/x86_64-linux-gnu/8/include -I./arch/x86/include -I./arch/x86/include/generated -I.$
make[2]: *** [scripts/Makefile.build:290: /var/lib/dkms/nvidia-current/418.74/build/nvidia/nv_uvm_interface.o] Error 1
make[2]: *** Waiting for unfinished jobs....I found this which seems to indicate this might be a kernel build issue:
https://github.com/clearlinux/distribution/issues/1306
Specifically this: https://github.com/clearlinux/distribution/issues/1306#issuecomment-547306578
