PDA

View Full Version : Trying to disable nouveau



Driftwould
31st July 2017, 11:45 PM
I have installed Fedora 26 on a laptop, ASUS ROG GL552VX-DM151D

Processor
CPU: Intel Core i7-6700HQ (2.60 GHz, 6 MB L3 Cache, up to 3.50 GHz)
Chipset: Mobile Intel Express Chipset
Graphic System
Graphic Card: NVIDIA GeForce GTX 950M (4GB GDDR5)


I have a few problems that I think might be related to nouveau.


1 - When I start up I get a few lines before the fedora start splash screen comes up. At shut down the same lines are still there once the gui closes and the computer never shuts down, I have too press the power button until it shuts down.





[ 1.889454] nouveau 0000:01:00.0: bus: MMIO read of 00000000 FAULT at 10ac08
[ IBUS]
[ 1.889511] nouveau 0000:01:00.0: priv: HUB0: 10ecc0 ffffffff (1e40822c)
[ 2.550584] nouveau 0000:01:00.0: DRM: Pointer to TMDS table invalid
[ 2.550588] nouveau 0000:01:00.0: DRM: Pointer to flat panel table invalid


2 - Also when I tried to go to settings, it hung up then crashed, now it will not open at all.

I found a website to remove nouveau and install the nvidia drivers. The first step was to check if my card is supported. I tried in the following as su -


lspci |grep -E "VGA|3D"

which just hangs up, then tried just


lspci

then


lshw

none of them worked.

Any suggestions on what to try next.
Thanks

dswaner
1st August 2017, 01:01 AM
You could try turning off nouveau acceleration by specifying "nouveau.noaccel=1" on the boot line. That might at least allow you to proceed with whatever you're going to do next.

smr54
1st August 2017, 01:55 AM
Doesn't rpmfusion have the NVidia drivers, which, when installed, disable the Nouveau ones?
https://rpmfusion.org/Howto/NVIDIA

Driftwould
1st August 2017, 03:29 AM
Doesn't rpmfusion have the NVidia drivers, which, when installed, disable the Nouveau ones?
https://rpmfusion.org/Howto/NVIDIA


I added the rpmfusion repository no problems




[root@europa ~]# dnf repolist
Last metadata expiration check: 0:23:15 ago on Mon 31 Jul 2017 10:01:30 PM EDT.
repo id repo name status
*fedora Fedora 26 - x86_64 53,912
*rpmfusion-free RPM Fusion for Fedora 26 - Free 536
*rpmfusion-free-updates RPM Fusion for Fedora 26 - Free - Updates 54
*rpmfusion-nonfree RPM Fusion for Fedora 26 - Nonfree 202
*rpmfusion-nonfree-updates RPM Fusion for Fedora 26 - Nonfree - Updates 3
*updates Fedora 26 - x86_64 - Updates 5,275





when I try to determine the card model I have the same problem that it just hangs up in the terminal window.



[root@europa ~]# /sbin/lspci | grep -e VGA


and



[root@europa ~]# /sbin/lspci | grep -e 3D


neither one returns any results.

Is there a problem with my shell or is it a different problem?

kldixon
1st August 2017, 09:33 AM
If you are just concerned about determining the card model then see:
https://nouveau.freedesktop.org/wiki/CodeNames/
You have:
NVIDIA GeForce GTX 950M = NV117 (GM107), NV110 family (Maxwell)
And from the 'supported chips' link of the rpmfusion howto
GeForce GTX 960M, PCI ID=139B
But the rpmfusion howto does indicate the driver required for GPUs later than 2010
It is a little worrying that lspci hangs. Does 'lspci -n' show anything?
On my machine:


# lspci -n
00:00.0 0600: 8086:0c00 (rev 06)
00:01.0 0604: 8086:0c01 (rev 06)
00:14.0 0c03: 8086:8c31 (rev 04)
00:16.0 0780: 8086:8c3a (rev 04)
00:19.0 0200: 8086:153b (rev 04)
00:1a.0 0c03: 8086:8c2d (rev 04)
00:1b.0 0403: 8086:8c20 (rev 04)
00:1d.0 0c03: 8086:8c26 (rev 04)
00:1f.0 0601: 8086:8c4a (rev 04)
00:1f.2 0106: 8086:8c02 (rev 04)
00:1f.3 0c05: 8086:8c22 (rev 04)
01:00.0 0300: 10de:0a65 (rev a2)
01:00.1 0403: 10de:0be3 (rev a1)

and my GPU is GeForce 210, PCI ID=0a65, which is supported by the 340.xx driver, by the 'supported chips' link.

However, you should probably use Negativo17.org rather than rpmfusion
https://negativo17.org/nvidia-driver-improvements-for-fedora-25/
https://blogs.gnome.org/uraeus/2016/11/01/discrete-graphics-and-fedora-workstation-25/
http://hansdegoede.livejournal.com/

And this is interesting, if a little old
http://www.thelinuxrain.com/articles/the-state-of-nvidia-optimus-on-linux

Driftwould
1st August 2017, 03:13 PM
It is a little worrying that lspci hangs. Does 'lspci -n' show anything?



It worries me too. I tried "lspci -n" and it still hangs up. After a few minutes I have to kill the xterminal window.:confused:

I'm I missing a package for bash?

kldixon
1st August 2017, 06:40 PM
The only thing I can think of is that your sysfs is not mounted.
Try chasing the directory tree of /sys to see if you can find the data lspci is using.
For example:


$ ls /sys/class/pci_bus
0000:00 0000:01
$ ls /sys/class/pci_bus/0000:01
cpuaffinity cpulistaffinity device power rescan subsystem uevent
$ ls /sys/class/pci_bus/0000:01/device
0000:00:01.0:pcie008 device local_cpus rescan
0000:01:00.0 dma_mask_bits modalias resource
0000:01:00.1 driver msi_bus revision
broken_parity_status driver_override msi_irqs subsystem
class enable numa_node subsystem_device
config firmware_node pci_bus subsystem_vendor
consistent_dma_mask_bits irq power uevent
d3cold_allowed local_cpulist remove vendor
$ ls /sys/class/pci_bus/0000:01/device/0000:01:00.0
boot_vga graphics i2c-9 resource1
broken_parity_status hwmon irq resource1_wc
class i2c-0 local_cpulist resource3
config i2c-1 local_cpus resource3_wc
consistent_dma_mask_bits i2c-10 modalias resource5
d3cold_allowed i2c-11 msi_bus revision
device i2c-2 msi_irqs rom
dma_mask_bits i2c-3 numa_node subsystem
driver i2c-4 power subsystem_device
driver_override i2c-5 remove subsystem_vendor
drm i2c-6 rescan uevent
enable i2c-7 resource vendor
firmware_node i2c-8 resource0
$ cat /sys/class/pci_bus/0000:01/device/0000:01:00.0/vendor
0x10de
$ cat /sys/class/pci_bus/0000:01/device/0000:01:00.0/device
0x0a65

Driftwould
2nd August 2017, 03:33 AM
Try chasing the directory tree of /sys to see if you can find the data lspci is using.



Thanks for the help. I'm not sure what I'm looking. I see you have a file called "boot_vga" where I do not. I went through the other directories looking for any thing with lspci. Below is part of the tracing of the /sys.




# ls /sys/class/pci_bus
0000:00 0000:01 0000:02 0000:03

]# ls /sys/class/pci_bus/0000:01/device
0000:00:01.0:pcie008 d3cold_allowed firmware_node msi_irqs resource vendor
0000:01:00.0 device irq numa_node revision
broken_parity_status dma_mask_bits local_cpulist pci_bus subsystem
class driver local_cpus power subsystem_device
config driver_override modalias remove subsystem_vendor
consistent_dma_mask_bits enable msi_bus rescan uevent

# ls /sys/class/pci_bus/0000:01/device/0000:01:00.0
broken_parity_status enable i2c-5 msi_irqs resource3
class firmware_node i2c-6 numa_node resource3_wc
config hwmon i2c-7 power resource5
consistent_dma_mask_bits i2c-10 i2c-8 remove revision
d3cold_allowed i2c-11 i2c-9 rescan rom
device i2c-12 irq reset subsystem
dma_mask_bits i2c-13 local_cpulist resource subsystem_device
driver i2c-14 local_cpus resource0 subsystem_vendor
driver_override i2c-15 modalias resource1 uevent
drm i2c-16 msi_bus resource1_wc vendor

# cat /sys/class/pci_bus/0000:01/device/0000:01:00.0/vendor
0x10de

# cat /sys/class/pci_bus/0000:01/device/0000:01:00.0/device
0x139a


does any of this look like data for lscpci?

antikythera
2nd August 2017, 07:30 AM
have you tried the same hardware diagnostics with live media rather than the actual installation? the output should be the same without the hanging.

kldixon
2nd August 2017, 09:19 AM
I was trying to determine why lspci hangs. As far as I understand it, lspci just parses the sysfs filesystem to acquire information and uses the hardware data in /usr/share/hwdata/pci.ids to transform that information into a more readable form. I first wondered if that file might not be present, which is why I asked you to try 'lspci -n'. You can try looking at pci.ids (or see: http://mj.ucw.cz/sw/pciutils/).

Now it is clear that sysfs is mounted so I have no idea why lspci is hanging.
(see: https://en.wikipedia.org/wiki/Sysfs for a description of sysfs.)

You have found the PCI bus location of your Nvidia chip. If you look up vendor 10de in pci.ids you will find 'NVIDIA Corporation' and device 139a is 'GM107M [GeForce GTX 950M]'. (Not 'GTX 960M' which I wrongly quoted in my previous post)

Driftwould
2nd August 2017, 04:34 PM
have you tried the same hardware diagnostics with live media rather than the actual installation? the output should be the same without the hanging.

I just tried it with same USB live media I used for the install in live mode and still have the same problem.

I tried the live media on an older computer to test the USB stick and it seems to be fine. Below is the output from a computer I am NOT having problems with.



[liveuser@localhost-live ~]$ lspci
00:00.0 Host bridge: Intel Corporation 3rd Gen Core processor DRAM Controller (rev 09)
00:01.0 PCI bridge: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor PCI Express Root Port (rev 09)
00:14.0 USB controller: Intel Corporation 7 Series/C210 Series Chipset Family USB xHCI Host Controller (rev 04)
00:16.0 Communication controller: Intel Corporation 7 Series/C216 Chipset Family MEI Controller #1 (rev 04)
00:1a.0 USB controller: Intel Corporation 7 Series/C216 Chipset Family USB Enhanced Host Controller #2 (rev 04)
00:1b.0 Audio device: Intel Corporation 7 Series/C216 Chipset Family High Definition Audio Controller (rev 04)
00:1c.0 PCI bridge: Intel Corporation 7 Series/C216 Chipset Family PCI Express Root Port 1 (rev c4)
00:1c.1 PCI bridge: Intel Corporation 7 Series/C210 Series Chipset Family PCI Express Root Port 2 (rev c4)
00:1c.3 PCI bridge: Intel Corporation 7 Series/C216 Chipset Family PCI Express Root Port 4 (rev c4)
00:1c.4 PCI bridge: Intel Corporation 7 Series/C210 Series Chipset Family PCI Express Root Port 5 (rev c4)
00:1d.0 USB controller: Intel Corporation 7 Series/C216 Chipset Family USB Enhanced Host Controller #1 (rev 04)
00:1f.0 ISA bridge: Intel Corporation HM77 Express Chipset LPC Controller (rev 04)
00:1f.2 SATA controller: Intel Corporation 7 Series Chipset Family 6-port SATA Controller [AHCI mode] (rev 04)
00:1f.3 SMBus: Intel Corporation 7 Series/C216 Chipset Family SMBus Controller (rev 04)
01:00.0 VGA compatible controller: NVIDIA Corporation GK107M [GeForce GTX 660M] (rev a1)
01:00.1 Audio device: NVIDIA Corporation GK107 HDMI Audio Controller (rev a1)
03:00.0 Network controller: Qualcomm Atheros AR9485 Wireless Network Adapter (rev 01)
04:00.0 Ethernet controller: Qualcomm Atheros AR8151 v2.0 Gigabit Ethernet (rev c0)


Like I said this is NOT from the computer I am having problems, so I am confidant the install media is fine.

kinappy
11th August 2017, 10:44 AM
I'm having the exact same issue on an Asus rog gl552vw.

Older live media lspci is ok, new media + install and lspci is hanging

I tried the same live media on a Latitude E7450 and lspci is working fine there

EDIT:
Just to test I add rd.driver.blacklist=nouveau to my grub config and now I can run lspci