nvidia-detect reports Quadro K620 as Optimus card ?

Issues related to hardware problems
taylorkh
Posts: 422
Joined: 2010/11/24 15:08:33
Location: North Carolina, USA

nvidia-detect reports Quadro K620 as Optimus card ?

Postby taylorkh » 2017/03/19 14:41:28

My new computer is a Dell Precision T3620 workstation. i7-6700 CPU, 16 GB RAM, 240 GB M.2 PCIE 3.0 4 lane boot drive, Intel graphics and a PNY Quadro K620 video card. I am running CentOS 7.3 + X windows system + Mate desktop environment. I have the following repos configured with yum-priorities:

CentOS Base & CR - priority=1
Elrepo - priority=5
EPEL - priority=10
nux-dextop - priority=20
scootersoftware - priority=25 (one program only - beyond-compare)

I installed the nvidia-detect package. It gives me this output

Code: Select all

[root@taylor20 Desktop]# nvidia-detect
kmod-nvidia
Optimus hardware detected: An Intel display controller was detected
Either disable the Intel display controller in the BIOS
or use the bumblebee driver to support Optimus hardware
This presents a couple of concerns which I will address shortly. I installed kmod-nvidia and the system works OK except for a couple of issues.

When I suspend and resume the computer I see this message on the black screen before the gui screen appears to ask me to authenticate

Code: Select all

[    1.115512] [drm:i915_gem_init_stolen [i915]] *ERROR* conflict detected with stolen region: [0xce000000 - 0xd0000000]
My research seems to point to this as an Intel video error. After I suspend and resume the computer a number of times I no longer observer the message. Instead when I resume I am presented with alternating displays of greeen and gray then red and white vertical stripes. These screens flash back and forth 11 times and then I see the gui authentication screen. I also see the message or the striped screens when I shut down or reboot the machine.

The second issue which may or may not be related - When I launch Firefox in a VMWare virtual machine (CentOS 7.3 or Ubuntu 16.04) the VM's display flashes once. This might be a Firefox phenomenon as I observer a sort of flicker in the upper right of the Firefox window when I invoke it on the CentOS 7.3 host. Sort of as if another window is launching but disappears into the background.

Here is where it get strange. The Quadro K620 is a workstation card. Optimus and the bumblebee driver are related to certain Nvidia cards on LAPTOPS for the purpose of saving power. I asked the card vendor, PNY and received this answer
I am not sure what the issue is with your intel display controller being detected but if you have it disabled in your BIOS and it is still being detected by CentOS you will need to speak to Dell for support regarding that matter or reach out to the CentOS community for assistance.

I can say that whatever version of nvidia-detect you are using is misreporting by telling you that you need a bumblebee optimus driver because those are laptop Nvidia GPUs, not desktop.


Dell "technical support" has been working for 3 weeks trying to explain to me how to disable the Intel video card in BIOS. They do not seem to have a clue.

I installed the bumblebee driver (after making a backup of the OS of course) but it made no improvement. In fact I lost the capability to accelerate 3D graphics on VMWare guests and I received several lines of error messages when doing a suspend/resume. I restored the OS and have not proceeded further down the bumblebee path.

A this point my questions are:

1 - Can anyone comment on the alleged error with nvidia-detect?
2 - While researching the bumblebee driver I recall something about being able to disable one of the GPUs. I did not book mark it and have not found it again. I see that there is a dependent package called kmod-bbswitch. Does anyone have experience with that program? Could it perhaps be used to disable the Intel GPU as it does not seem to be possible in BIOS?
3 - PNY also recommended that I install the Nvidia driver myself from nvidia.com. I have had GREAT luck with drivers from Elrepo on CentOS 6 for the past 7 years so I am reluctant to "roll my own." Is this worth pursuing?

I have recorded the output from getinfo.sh driver below for any information which that might provide.

TIA,

Ken

Code: Select all

Information for driver problems.
== BEGIN uname -rmi ==
3.10.0-514.10.2.el7.x86_64 x86_64 x86_64
== END   uname -rmi ==

== BEGIN rpm -qa \*-release\* ==
elrepo-release-7.0-2.el7.elrepo.noarch
epel-release-7-9.noarch
nux-dextop-release-0-5.el7.nux.noarch
centos-release-7-3.1611.el7.centos.x86_64
== END   rpm -qa \*-release\* ==

== BEGIN cat /etc/redhat-release ==
CentOS Linux release 7.3.1611 (Core)
== END   cat /etc/redhat-release ==

== BEGIN getenforce ==
Disabled
== END   getenforce ==

== BEGIN free -m ==
              total        used        free      shared  buff/cache   available
Mem:          15778         999         186          87       14591       14316
Swap:             0           0           0
== END   free -m ==

== BEGIN rpm -qa kernel\* | sort ==
kernel-3.10.0-327.el7.x86_64
kernel-3.10.0-514.10.2.el7.x86_64
kernel-3.10.0-514.6.1.el7.x86_64
kernel-3.10.0-514.6.2.el7.x86_64
kernel-devel-3.10.0-514.10.2.el7.x86_64
kernel-devel-3.10.0-514.6.1.el7.x86_64
kernel-devel-3.10.0-514.6.2.el7.x86_64
kernel-headers-3.10.0-514.10.2.el7.x86_64
kernel-tools-3.10.0-514.10.2.el7.x86_64
kernel-tools-libs-3.10.0-514.10.2.el7.x86_64
== END   rpm -qa kernel\* | sort ==

== BEGIN lspci -nn ==
00:00.0 Host bridge [0600]: Intel Corporation Skylake Host Bridge/DRAM Registers [8086:191f] (rev 07)
00:01.0 PCI bridge [0604]: Intel Corporation Skylake PCIe Controller (x16) [8086:1901] (rev 07)
00:02.0 Display controller [0380]: Intel Corporation HD Graphics 530 [8086:1912] (rev 06)
00:14.0 USB controller [0c03]: Intel Corporation Sunrise Point-H USB 3.0 xHCI Controller [8086:a12f] (rev 31)
00:14.2 Signal processing controller [1180]: Intel Corporation Sunrise Point-H Thermal subsystem [8086:a131] (rev 31)
00:16.0 Communication controller [0780]: Intel Corporation Sunrise Point-H CSME HECI #1 [8086:a13a] (rev 31)
00:17.0 SATA controller [0106]: Intel Corporation Sunrise Point-H SATA controller [AHCI mode] [8086:a102] (rev 31)
00:1b.0 PCI bridge [0604]: Intel Corporation Sunrise Point-H PCI Root Port #17 [8086:a167] (rev f1)
00:1c.0 PCI bridge [0604]: Intel Corporation Sunrise Point-H PCI Express Root Port #1 [8086:a110] (rev f1)
00:1f.0 ISA bridge [0601]: Intel Corporation Sunrise Point-H LPC Controller [8086:a149] (rev 31)
00:1f.2 Memory controller [0580]: Intel Corporation Sunrise Point-H PMC [8086:a121] (rev 31)
00:1f.3 Audio device [0403]: Intel Corporation Sunrise Point-H HD Audio [8086:a170] (rev 31)
00:1f.4 SMBus [0c05]: Intel Corporation Sunrise Point-H SMBus [8086:a123] (rev 31)
00:1f.6 Ethernet controller [0200]: Intel Corporation Ethernet Connection (2) I219-LM [8086:15b7] (rev 31)
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GM107GL [Quadro K620] [10de:13bb] (rev a2)
01:00.1 Audio device [0403]: NVIDIA Corporation Device [10de:0fbc] (rev a1)
02:00.0 Non-Volatile memory controller [0108]: Device [1987:5007] (rev 01)
03:00.0 PCI bridge [0604]: Texas Instruments XIO2001 PCI Express-to-PCI Bridge [104c:8240]
04:02.0 USB controller [0c03]: VIA Technologies, Inc. VT82xx/62xx UHCI USB 1.1 Controller [1106:3038] (rev 62)
04:02.1 USB controller [0c03]: VIA Technologies, Inc. VT82xx/62xx UHCI USB 1.1 Controller [1106:3038] (rev 62)
04:02.2 USB controller [0c03]: VIA Technologies, Inc. USB 2.0 [1106:3104] (rev 65)
== END   lspci -nn ==

== Warning: lsusb is not installed ==

== BEGIN rpm -qa kmod\* kmdl\* ==
kmod-libs-20-9.el7.x86_64
kmod-20-9.el7.x86_64
kmod-nvidia-375.39-2.el7.elrepo.x86_64
== END   rpm -qa kmod\* kmdl\* ==

NedSlider
Forum Moderator
Posts: 2886
Joined: 2005/10/28 13:11:50
Location: UK

Re: nvidia-detect reports Quadro K620 as Optimus card ?

Postby NedSlider » 2017/03/20 07:18:40

Please could you show the verbose output from nvidia-detect with -v:

nvidia-detect -v

Thanks

NedSlider
Forum Moderator
Posts: 2886
Joined: 2005/10/28 13:11:50
Location: UK

Re: nvidia-detect reports Quadro K620 as Optimus card ?

Postby NedSlider » 2017/03/20 07:29:59

taylorkh wrote:My new computer is a Dell Precision T3620 workstation. i7-6700 CPU, 16 GB RAM, 240 GB M.2 PCIE 3.0 4 lane boot drive, Intel graphics and a PNY Quadro K620 video card.

Code: Select all

== BEGIN lspci -nn ==
00:02.0 Display controller [0380]: Intel Corporation HD Graphics 530 [8086:1912] (rev 06)
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GM107GL [Quadro K620] [10de:13bb] (rev a2)
== END   lspci -nn ==



nvidia-detect scans the pci bus and reports on the video controllers it finds. In your case, the same as the output from lspci above, it finds both Intel AND Nvidia controllers are present. Because it finds both Intel AND Nvidia controllers, it assumes this is Optimus hardware in a laptop.

I have a i3-6100 CPU based system with built in HD Graphics 530, and when I plug in a discrete nvidia card, the on chip Intel controller is disabled and does NOT appear in the output of lspci. This is what is causing the issue.

Ignore the Optimus / Bumblebee advise from nvidia-detect, this is for laptop setups. Uninstall the bumblebee driver. Just install your nvidia drivers of choice and you should be good to go. If there is an option to disable the on chip Intel controller in the bios, make sure it is disabled.

taylorkh
Posts: 422
Joined: 2010/11/24 15:08:33
Location: North Carolina, USA

Re: nvidia-detect reports Quadro K620 as Optimus card ?

Postby taylorkh » 2017/03/20 12:42:28

Thanks NedSlider,

As requested

Code: Select all

[root@taylor20 Desktop]# nvidia-detect -v
Probing for supported NVIDIA devices...
[10de:13bb] NVIDIA Corporation GM107GL [Quadro K620]
This device requires the current 375.39 NVIDIA driver kmod-nvidia
[8086:1912] Intel Corporation HD Graphics 530
Optimus hardware detected: An Intel display controller was detected
Either disable the Intel display controller in the BIOS
or use the bumblebee driver to support Optimus hardware
Your explanation makes sense. As I think I mentioned in the original I restored the OS to the state before playing with bumblebee (Clonezilla - fantastic tool). As to the BIOS... When I plug in the Nvidia card I see these options under video:

Primary display
( ) Auto
( ) Intel
( ) Nvidia

and a note to the effect "Choosing anything other than Auto will enable the on-board graphics and you will have to re plug in your monitor" - whatever that is supposed to mean. Tech support can not tell me. The only other graphics related choice is [ ] Enable multiple displays Note: "Only applies to Windows 7 and above. Does not apply to other operating systems."

I have I think tried all possible permutations of these settings (except choosing Intel as primary - that would probably work :roll:

Thanks again,

Ken

p.s. Some years ago I had a VISA card with an RFID chip (not the current security chip). I did not want an RFID chip due to security concerns. I called the bank and told them I wanted it disabled on my account. They said they couldn't do that. I told them to send me a card without the chip. They said they would not do that. They told me I had to have the RFID chip. The location of the chip was obvious in the card. I took it to my shop and used a gasket punch to remove a 1/2" diameter piece of the card containing the chip. The card still worked fine. Perhaps If I determine where the Intel on-board graphics are located on the CPU chip I can just drill it out. I bet that would disable it :mrgreen:

User avatar
jlehtone
Posts: 1857
Joined: 2007/12/11 08:17:33
Location: Finland

Re: nvidia-detect reports Quadro K620 as Optimus card ?

Postby jlehtone » 2017/03/20 13:38:15

taylorkh wrote:As to the BIOS... When I plug in the Nvidia card I see these options under video:

Primary display
( ) Auto
( ) Intel
( ) Nvidia

and a note to the effect "Choosing anything other than Auto will enable the on-board graphics and you will have to re plug in your monitor" - whatever that is supposed to mean. Tech support can not tell me.

I know two Dell instances:

Code: Select all

#### Dell OptiPlex 9020 ####
00:02.0 VGA compatible controller [0300]: Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor Integrated Graphics Controller [8086:0412] (rev 06)
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GM107GL [Quadro K620] [10de:13bb] (rev a2)

#### Dell Precision Tower 3620 ####
00:02.0 VGA compatible controller [0300]: Intel Corporation HD Graphics 530 [8086:1912] (rev 06)
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GM107GL [Quadro K620] [10de:13bb] (rev a2)

... and I simply ignore the bumble-messages; the systems are functional.

I do not recall seeing any such "video" option within BIOS/EFI (I may not have looked hard enough.)


The issue is clearly about what the Dell BIOS does do and what it should do.

I have no idea what the BIOS should do (automatically or via option) with the IGP when a discrete card is added. In theory one could be interested in using both GPUs; one for graphics and other for compute. Laptops are different: there the GPUs clearly share the display connectors and both can send images to the shared display (with bumblebee).



I've seen a ~15 PC class. Identical desktops. Intel motherboard had IGP (Intel CPU did not have one yet back then). The manual clearly stated that the IGP is automatically disabled, if a discrete GPU is installed. It was true on all but one ...

taylorkh
Posts: 422
Joined: 2010/11/24 15:08:33
Location: North Carolina, USA

Re: nvidia-detect reports Quadro K620 as Optimus card ?

Postby taylorkh » 2017/03/20 14:33:50

Thanks again jlehtone,

I have been using Dell computers since they were called PCs Ltd. The last couple of years have been a fiasco. I purchased an XPS with an i7-6700 during the Black Friday/Cyber Monday madness in 2015. I could not get any Linux distro to install. Sent it back. Since then I have found information on some tweaks which are supposed to resolve the issue (or a newer kernel which recognizes the hardware).

Then last year I came across a "deal" on a T20 server - just an empty box, no HD no OS for $129. That was less than Dell wanted for a 1 year extension on the warranty on the T20 I already had. So I ordered it. Normally these pre-configured packages ship quick. This one showed a ship date about 2 weeks out. Strange. A few days after placing the order I got a phone call from someone purporting to be from Dell "we want to cancel your order, is that OK?" No I said. I watch the order status day by day and it seemed to be OK. The shipping date even advanced by one day. The day before it was due to ship it showed canceled.

I called Dell and asked WTF? I was told the computer was "end of life" which in Dell speak means that the no longer offer that particular configuration. When asked what they had to offer to substitute I was offered a T130 for about $550. That sort of defeated the purpose of having a cheap spare server so I thought I would wait for the T20 to pop up again (and is has several times). But then I got a little irritated and filed a complaint with the Better Business Bureau. The day after Bell received the BBB complaint I received a phone call from someone in Dell corporate . Bottom line they sold me a T130 for $129! It dawned on me that what the had done could be considered a "bait and switch" which is illegal. I guess that is why they reposnded to the BBB complaint so quickly.

The T130 arrived in 2 days. Nice box. A neat wiring harness for 4 hard drives instead of a bunch of individual data and power cables. Except - there was no place on the system board to plug in the end of the harness. It was the harness for a PERC raid card which I did not have, nor want. It took the "premier tech support" dude almost an hour to figure out that I needed the "other" harness. There are only two available for the T130. They did overnight the replacement harness.

When I ordered the Precision workstation I picked a machine on which Dell offered Ubuntu or RHEL figuring I would not have any compatibility issues. Actually I ordered it with Ubuntu 14.04 as that saved me $93 by not having a Windoze 10 license. I spent 2 weeks getting the machine configured with CentOS 7.3 and tweaked to my liking. I was about ready to do a final data transfer and replace my old CentOS 6 machine when I began getting kernel crashes when I suspended the thing. The message implied something corrupt in BIOS related to power management. I looked in BIOS and did not see anything suspicious related to power management. I had not changed anything in that area. Having nothing better to try I did a "Restore to factory" process. That fixed the crash issue. However, NOTHING in BIOS changed from what I had changed.

Being curious about the restore which did not restore anything I initiated a chat with tech support. At the end of 4 hours tech support had managed to render the machine such that it would not power off. The support rep said "I need to refer this to advanced support". No, I said. your need to refer this to the RMA department. I have had it less than 30 days. You can have it back. So it went back and I ordered the one I have now.

Tech support can not tell me how to disable the on-board Intel video, how to reset the BIOS to the state it came from the factory, if/how I am supposed to connect a floppy drive to the machine - it does have a boot from floppy option in BIOS, nor about anything else. DD-WRT and Open-WRT allow router firmware to be upgraded to something more useful. I wonder if anyone has open source upgrades for PC BIOS? :mrgreen:

I guess I should have ordered a WihisperStation - Tesla from microway.com. 44 processor cores, 1 TB of RAM and 4 Tesla GPUs. I bet that would load Firefox in less than 20 seconds - well probably not.

Ken

NedSlider
Forum Moderator
Posts: 2886
Joined: 2005/10/28 13:11:50
Location: UK

Re: nvidia-detect reports Quadro K620 as Optimus card ?

Postby NedSlider » 2017/03/20 19:22:47

Thank you both for feedback.

I think I'll revise the wording of the notification in nvidia-detect when both Intel and NVIDIA video controllers are detected. Any suggestions as to what might be more suitable? Previously I had just assumed that if both hardware was present (visible) then it must be an Optimus/laptop setup, but clearly that isn't always true.

Question: Out of curiosity, do both controllers work (separately) if a monitor is hooked up?

User avatar
TrevorH
Forum Moderator
Posts: 20996
Joined: 2009/09/24 10:40:56
Location: Brighton, UK

Re: nvidia-detect reports Quadro K620 as Optimus card ?

Postby TrevorH » 2017/03/20 22:14:57

Is there such a thing as an optimus desktop? If not then I'd suggest just amending the wording to include "If you are on a laptop with ..."
CentOS 5 died in March 2017 - migrate NOW!
Full time Geek, part time moderator. Use the FAQ Luke

taylorkh
Posts: 422
Joined: 2010/11/24 15:08:33
Location: North Carolina, USA

Re: nvidia-detect reports Quadro K620 as Optimus card ?

Postby taylorkh » 2017/03/20 22:42:54

To TrevorH - I have no idea if there is an Optimus desktop. From what I read about bumblebee it seems that the whole Optimus concept is for saving energy to extend battery life. Your proposed wording would make sense to me.

To NedSlid - Sorry about the Dell rant. This has all been very frustrating. I will put together a little test plan to see what happens when I plug in different combinations of monitors into different ports. Perhaps I can get 5 monitors running instead of just 2 :D That said...

I mentioned the issue with the flashing vertical lines when I wake the machine or, if they are present when waking, they will show when shutting down. In the later case I have no idea if the machine will power off. I get impatient and hit the button after a couple of minutes. But it gets stranger...

I discovered last evening, totally by accident, that if i hit a key on the keyboard when the computer is suspended it will awake. When I do this, and I have quite a few times since the last reboot, the message displays but not the flashing lines. I have probably jinxed myself by stating this :( I also discovered that moving the mouse will wake the machine as will a static discharge to the CASE! I discovered that when I picked up my wallet which was sitting on top of the computer. The thing is SUPPOSED to be grounded via the power cord. To further complicate matters I have vlc running - which suspend scolds me about. Perhaps that is what banishes the flashing lines.

I will post back in the next day or so after I have done some monitor testing.

Thanks again for your input.

Ken

taylorkh
Posts: 422
Joined: 2010/11/24 15:08:33
Location: North Carolina, USA

Re: nvidia-detect reports Quadro K620 as Optimus card ?

Postby taylorkh » 2017/03/20 22:55:08

Here is a little teaser regarding GPU vs GPU. From the tail end of dmesg

Code: Select all

[    3.370627] snd_hda_intel 0000:00:1f.3: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])
[    3.380152] snd_hda_intel 0000:00:1f.3: irq 135 for MSI/MSI-X
[    3.399466] snd_hda_codec_realtek hdaudioC0D0: autoconfig for ALC3861: line_outs=1 (0x14/0x0/0x0/0x0/0x0) type:line
[    3.399467] snd_hda_codec_realtek hdaudioC0D0:    speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[    3.399467] snd_hda_codec_realtek hdaudioC0D0:    hp_outs=1 (0x1b/0x0/0x0/0x0/0x0)
[    3.399468] snd_hda_codec_realtek hdaudioC0D0:    mono: mono_out=0x0
[    3.399468] snd_hda_codec_realtek hdaudioC0D0:    inputs:
[    3.409366] input: HDA Intel PCH Headphone Mic as /devices/pci0000:00/0000:00:1f.3/sound/card0/input9
[    3.409414] input: HDA Intel PCH Line Out as /devices/pci0000:00/0000:00:1f.3/sound/card0/input10
[    3.409458] input: HDA Intel PCH HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input11
[    3.409479] input: HDA Intel PCH HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input12
[    3.409498] input: HDA Intel PCH HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input13
[    3.475184] input: HDA NVidia HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input14
[    3.475214] input: HDA NVidia HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input15
[    3.563836] EXT4-fs (nvme0n1p5): mounted filesystem with ordered data mode. Opts: discard
[    3.566322] EXT4-fs (nvme0n1p1): mounted filesystem with ordered data mode. Opts: discard
[    3.574736] EXT4-fs (sdc1): mounted filesystem with ordered data mode. Opts: discard
[    3.593919] type=1305 audit(1489917926.381:3): audit_pid=622 old=0 auid=4294967295 ses=4294967295 res=1
A few lines above is the error message

Code: Select all

[    1.130608] [drm:i915_gem_init_stolen [i915]] *ERROR* conflict detected with stolen region: [0xce000000 - 0xd0000000]
[    1.130779] [drm] Memory usable by graphics device = 4096M
[    1.130784] [drm] Replacing VGA console driver
[    1.133499] Console: switching to colour dummy device 80x25
[    1.139736] i915 0000:00:02.0: irq 133 for MSI/MSI-X
[    1.140068] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[    1.140070] [drm] Driver supports precise vblank timestamp query.
[    1.140962] [drm] Finished loading i915/skl_dmc_ver1.bin (v1.26)
and prior to that

Code: Select all

[    0.832174] Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
[    0.832182] nvidia: loading out-of-tree module taints kernel.
[    0.832184] nvidia: module license 'NVIDIA' taints kernel.
[    0.832185] Disabling lock debugging due to kernel taint
[    0.834489] nvidia: module verification failed: signature and/or required key missing - tainting kernel
[    0.837337] vgaarb: device changed decodes: PCI:0000:01:00.0,olddecodes=io+mem,decodes=none:owns=io+mem
[    0.837403] nvidia-nvlink: Nvlink Core is being initialized, major device number 245
[    0.837409] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  375.39  Tue Jan 31 20:47:00 PST 2017 (using threaded interrupts)
[    0.838785] Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
[    0.839856] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms  375.39  Tue Jan 31 19:41:48 PST 2017
[    0.840377] Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
[    0.840990] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
I am not running secure boot. Could that have something to do with the Nvidia mischief I wonder? It looks like your kmod-nvidia is supposed to be signed?

Ken