kernel: NVRM: API mismatch: the client has the version 364.12, but\x0aNVRM: this kernel module has the version 361.28. Please\x0aNVRM: make sure that this kernel module and all NVIDIA driver\x0aNVRM: components have the same version. Also get this error when attempting to use lsmod -r nvidia-drivers "nvidia-uvm not found" To clear out the 364.12 module it is necessary to reboot. Yet even with 361.28 loaded, if lsmod -r nvidia-drivers is used, the same error as 364.12 "nvidia-uvm not found"
Err, lsmod doesn't even have a -r flag. Can you make an attempt to more carefully describe the problem you are having?
Not seeing any errors here: kernel: nvidia: module license 'NVIDIA' taints kernel. kernel: Disabling lock debugging due to kernel taint kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 246 kernal: NVRM: loading NVIDIA UNIX x86_64 Kernel Module 364.12 Wed Mar 16 21:11:26 PDT 2016 from my last boot of 4.5.0
Sorry my fault. I most certainly meant modprobe -r nvidia or modprobe -rf nvidia. I was in a bit of a hurry this morning, yes my fault for not being more careful. That was simply a typo when I typed this bug report. I have been a bit busy today so do not have much time, as this was on my workstation and have work to do. So simply masked 364.12. I do not get any other errors. 364.12 compiles fine. It is rather odd, that I am unable to remove the module. There have been a number of problems lately with all the current changes with the nvidia-drivers. I have had to mask and wait for fixes more times in the past few months, than I have in many years. Wish I could provide more at this time. I need to get some things completed. In the next day or two, will get back to it, and see if I can find some more details. Xorg.log provides nothing other than check kernel log, which is what I posted from the kernel log. Going to compile 364.12 again and take a look at the build.log. I am compiling with distcc cluster. Sometimes there are problems with distributed compiling, though in most cases if there is an issue, the compile fails. While in this case, the compile completes just fine.
For some reason modprobe -r nvidia-drivers is now working without errors. No errors in kernel log, removing or adding nvidia and nvidia-modeset modules. Rebooted today more times than in the past 2 years. Might have cleared something out. Here are the use flags: USE="X acpi driver gtk3 kms tools -compat -multilib -pax_kernel -static-libs -uvm" Xorg.log: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/Xorg.0.log dmesg: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/dmesg kernel-messages-edited: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/kernel-messages make.conf: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/make.conf first-compile-log: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/x11-drivers%3Anvidia-drivers-364.12%3A20160324-031829.log second-compile-log: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/x11-drivers%3Anvidia-drivers-364.12%3A20160324-031832.log third-compile-log: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/x11-drivers%3Anvidia-drivers-364.12%3A20160324-033239.log One thing I have noticed with 364.12 is that the screen does not flicker with the automatic switch from xorg-x11 to nvidia. It stays on xorg-x11 unlike recent previous versions that automatically select nvidia. Does not matter if I manually set nvidia. I get that upstream is attempting to automate things, but personaly I would rather manually configure everything. I apologize for not providing more details in the first post. The api mismatch tells me something, but perhaps I am mistaken. I have had about 4 hours sleep in the past 2 days. Dealing with the amazon aws linux ami upgrade bump, been busy updating server clusters. For some reason when I re-read my first post, the lsmod rather than modprode did not register. Your are welcome to beat about the head and shoulders till I fall asleep! : ) If there is anything else that could be of help. Simply ask and I will do my best to provide it.
(In reply to Jeff Gerhardt from comment #4) > For some reason modprobe -r nvidia-drivers is now working without errors. No > errors in kernel log, removing or adding nvidia and nvidia-modeset modules. > Rebooted today more times than in the past 2 years. Might have cleared > something out. > > Here are the use flags: > USE="X acpi driver gtk3 kms tools -compat -multilib -pax_kernel -static-libs > -uvm" > > Xorg.log: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/Xorg.0.log > dmesg: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/dmesg > kernel-messages-edited: > https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/kernel-messages > make.conf: https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/make.conf > first-compile-log: > https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/x11-drivers%3Anvidia- > drivers-364.12%3A20160324-031829.log > second-compile-log: > https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/x11-drivers%3Anvidia- > drivers-364.12%3A20160324-031832.log > third-compile-log: > https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/x11-drivers%3Anvidia- > drivers-364.12%3A20160324-033239.log > > One thing I have noticed with 364.12 is that the screen does not flicker > with the automatic switch from xorg-x11 to nvidia. > It stays on xorg-x11 unlike recent previous versions that automatically > select nvidia. > > Does not matter if I manually set nvidia. I get that upstream is attempting > to automate things, but personaly I would rather manually configure > everything. > > I apologize for not providing more details in the first post. The api > mismatch tells me something, but perhaps I am mistaken. I have had about 4 > hours sleep in > the past 2 days. Dealing with the amazon aws linux ami upgrade bump, been > busy updating server clusters. For some reason when I re-read my first post, > the lsmod rather than modprode did not register. Your are welcome to beat me > about the head and shoulders till I fall asleep! :) > > If there is anything else that could be of help. Simply ask and I will do my > best to provide it.
I have attempted compiling with and without distcc, added flags, removed flags, always the same result. The modprobe -r nvidia is once again giving the nvidia-uvm not found error. Set the uvm flag, same result. I forgot to add my xorg.conf here it is. Xorg will not run without that configuration. Been using it for about 2.5 years. Without having to modify it for any previous versions. https://s3-us-west-2.amazonaws.com/gentoo-compile-logs/xorg.conf
the API mismatch error is 100% expected and not a bug can't really explain the nvidia-uvm part. you should post the output of `lsmod` before you run modprobe.
Prior to using startx it is: nvidia After xorg is started: nvidia_modeset nvidia I do not use a display manager, while nvidia drivers are the only module, everything else is compiled into the kernel. Old habits die hard, been doing it this way since 1999. LFS rules! 361.28-r2 works fine. While other than the occasional release, nearly every other prior version has worked fine since nvidia-drivers was first stabilized. As I have previously stated, up until the recent releases, this past few months. Have not had a problem for many years. Not a big deal to have to occasionally mask a version. The only reason I have posted. Is because there has been a problem with most versions this past few months not working. Simply attempting, to bring light to the issue, so as to prevent future difficulties with bugs continuing to creep into future so called stable releases.
Have you fixed this? I'm resolving as NEEDINFO. Reopen if the problem persists and I will assign to nvidia gentoo guys asap.
*** Bug 679310 has been marked as a duplicate of this bug. ***
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=9fd6d358a4e831724c79f9369c3c86dea00cddd0 commit 9fd6d358a4e831724c79f9369c3c86dea00cddd0 Author: Jeroen Roovers <jer@gentoo.org> AuthorDate: 2019-03-03 13:09:53 +0000 Commit: Jeroen Roovers <jer@gentoo.org> CommitDate: 2019-03-03 13:11:05 +0000 x11-drivers/nvidia-drivers: USE=uvm: add udev rule, fix rmmod Package-Manager: Portage-2.3.62, Repoman-2.3.12 Bug: https://bugs.gentoo.org/506696 Bug: https://bugs.gentoo.org/578126 Signed-off-by: Jeroen Roovers <jer@gentoo.org> .../nvidia-drivers/files/nvidia-uvm.udev-rule | 1 + .../nvidia-drivers-390.116-r1.ebuild | 589 ++++++++++++++++++++ .../nvidia-drivers-410.104-r1.ebuild | 592 +++++++++++++++++++++ .../nvidia-drivers/nvidia-drivers-415.27-r1.ebuild | 592 +++++++++++++++++++++ .../nvidia-drivers/nvidia-drivers-418.43-r1.ebuild | 585 ++++++++++++++++++++ 5 files changed, 2359 insertions(+)
Fixed for me while upgrading from 418.43 to 418.56, thank you.
Seems was already fixed, closing.
