Hello.

It seems, that I'm experiencing exactly the same problem. During kernel update process to v4.4.0-143.169 (from '-proposed') there was an error message about some issues with 'nvidia' module etc. '/var/lib/dkms/nvidia-304/304.135/build/make.log' file contains a full build process informations. Here are some examples (I can paste it somewhere):


============[ NVIDIA Build log examples ]============

In file included from ./arch/x86/include/asm/preempt.h:5:0,
                 from include/linux/preempt.h:59,
                 from include/linux/spinlock.h:50,
                 from include/linux/seqlock.h:35,
                 from include/linux/time.h:5,
                 from include/uapi/linux/timex.h:56,
                 from include/linux/timex.h:56,
                 from include/linux/sched.h:19,
                 from include/linux/utsname.h:5,
                 from /var/lib/dkms/nvidia-304/304.135/build/nv-linux.h:40,
                 from /var/lib/dkms/nvidia-304/304.135/build/nv.c:13:
include/linux/percpu-refcount.h: In function ‘percpu_ref_get_many’:
./arch/x86/include/asm/percpu.h:130:31: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
          ((val) == 1 || (val) == -1)) ?  \
                               ^
(...)

include/linux/percpu-refcount.h: In function ‘percpu_ref_get_many’:
./arch/x86/include/asm/percpu.h:130:31: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
          ((val) == 1 || (val) == -1)) ?  \
                               ^
(...)

include/linux/percpu-defs.h:394:11: note: in expansion of macro ‘this_cpu_add_1’
   case 1: stem##1(variable, __VA_ARGS__);break;  \
           ^
(...)

include/linux/percpu-refcount.h:276:3: note: in expansion of macro ‘this_cpu_sub’
   this_cpu_sub(*percpu_count, nr);
   ^
(...)

include/linux/compiler.h:165:40: note: in definition of macro ‘likely’
 # define likely(x) __builtin_expect(!!(x), 1)
                                        ^
(...)

In file included from /var/lib/dkms/nvidia-304/304.135/build/nv-linux.h:82:0,
                 from /var/lib/dkms/nvidia-304/304.135/build/nv-mlock.c:15:
include/linux/mm.h:1222:6: note: declared here
 long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
      ^
(...)

make[2]: *** [_module_/var/lib/dkms/nvidia-304/304.135/build] Error 2
NVIDIA: left KBUILD.
nvidia.ko failed to build!
Makefile:257: commands for 'module' object failed
make[1]: *** [module] Error 1
makefile:59: commands for 'module' object failed
make: *** [module] Error 2


● NOTE: There is more definitions of, mentioned above, 'macros'. For example: ‘__pcpu_size_call’, ‘this_cpu_add’, ‘percpu_add_op’, ‘this_cpu_add_4’ etc. Generally, they all are repeated in build log.

============[ END of NVIDIA Build log example ]============


That's is only an example (whole build log contains about ~1500 lines). Most interesting and strange part is that I'm not using 'nvidia' or 'nouveau' module, but I still can log to system via LightDM! Why is that? Shouldn't there be some graphics issues? Here are some more informations about 'nvidia' module (NOTE: I'm logged in and using v4.4.0-143-generic kernel - on which 'nvidia' module failed to build):

[~]$ modinfo nvidia
modinfo: ERROR: Module nvidia not found.

[~]$ sudo modprobe nvidia
modprobe: ERROR: ../libkmod/libkmod-module.c:832 kmod_module_insert_module() could not find module by name='nvidia_304'
modprobe: ERROR: could not insert 'nvidia_304': Unknown symbol in module, or unknown parameter (see dmesg)

[~]$ sudo modprobe nvidia_304
modprobe: FATAL: Module nvidia_304 not found in directory /lib/modules/4.4.0-143-generic

(NOTE: dmesg(1) is empty - there is no informations about 'nvidia' issues).

[~]$ lsmod |grep nvidia
[~]$ lsmod |grep nouveau

By the way: there were some problems with graphics when monitor blacked out (there should 'light-locker' show up), because of - for example - no mouse movements or no reaction from user side and after resume colors were bright, fonts thiner etc. If it's about X.org logs, there is something like this in '/var/log/Xorg.0.log' file:

(II) LoadModule: "nvidia"
(II) Loading /usr/lib/i386-linux-gnu/xorg/extra-modules/nvidia_drv.so
(II) Module nvidia: vendor="NVIDIA Corporation"
compiled for 4.0.2, module version = 1.0.0
(...)
(EE) NVIDIA: Failed to load the NVIDIA kernel module. Please check your
(EE) NVIDIA:     system's kernel log for additional error messages.
(II) UnloadModule: "nvidia"
(II) Unloading nvidia
(EE) Failed to load module "nvidia" (module-specific error, 0)
(II) LoadModule: "nouveau"
(II) Loading /usr/lib/xorg/modules/drivers/nouveau_drv.so

Now, some technical informations about versions etc.:

✗ Linux kernel: v4.4.0-143-generic
✗ Description:	Ubuntu 16.04.5 LTS
  Release:	16.04

✗ nvidia-304 304.135-0ubuntu0.16.04.2 
✗ nvidia-304-updates 304.135-0ubuntu0.16.04.2
✗ xserver-xorg-video-nouveau 1:1.0.12-1build2

✗ lspci(8) info: 00:10.0 VGA compatible controller: NVIDIA Corporation C73 [GeForce 7100 / nForce 630i] (rev a2)

So, there is a problem with latest '-proposed' kernel and 'nvidia' module. However, everything seems to work okay: even when there is no graphics module loaded (vide no 'nvidia', no 'nouveau'). Strange.

Thanks, best regards.