# Cannot Install Tesla K40 driver on Ubuntu 18.04 **URL:** \ **Category:** Linux\ **Created:** [March 22, 2020, 8:38am UTC](https://forums.developer.nvidia.com/t/cannot-install-tesla-k40-driver-on-ubuntu-18-04/116386 "2020-03-22T08:38:08Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![ted4q](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@ted4q](https://forums.developer.nvidia.com/u/ted4q)\ **Post date:** [March 22, 2020, 8:38am UTC](https://forums.developer.nvidia.com/t/cannot-install-tesla-k40-driver-on-ubuntu-18-04/116386/1 "2020-03-22T08:38:08Z")
Motherboard is a GA-Z97X-GAMING G1 WIFI-BK from Gigabyte and I have 4 Tesla K40 installed side by side. All 4 are recognized with lspci | grep -i nvidia. So far I have attempted to install via: 1. Ubuntu Repository 2. Ubuntu graphics-drivers PPA 3. Driver .Deb 4. Cuda Toolkit .Deb The driver installs with all of these methods. Yet when I run nvidia-smi it gives me an error that the driver is not loaded. I have read that installing the driver via the toolkit runfile fixes the issue for some. However, the driver installation fails via this method. Here is an excerpt from nvidia-installer.log: ERROR: Unable to load the kernel module ‘nvidia.ko’. This happens most frequently when this kernel module was built against the wrong or improperly configured kernel sources, with a version of gcc that differs from the one used to build the target kernel, or if another driver, such as nouveau, is present and preve$ Please see the log entries ‘Kernel module load error’ and ‘Kernel messages’ at the end of the file ‘/var/log/nvidia-installer.log’ for more information. → Kernel module load error: No such device → Kernel messages: [97.527156] ipmi device interface [97.531411] nvidia: loading out-of-tree module taints kernel. [97.531416] nvidia: module license ‘NVIDIA’ taints kernel. [97.531416] Disabling lock debugging due to kernel taint [97.535064] nvidia: module verification failed: signature and/or required key missing - tainting kernel [97.539600] nvidia-nvlink: Nvlink Core is being initialized, major device number 238 [97.539882] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid: NVRM: BAR0 is 0M @ 0x0 (PCI:0000:03:00.0) [97.539882] NVRM: The system BIOS may have misconfigured your GPU. [97.539885] nvidia: probe of 0000:03:00.0 failed with error -1 [97.539919] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid: NVRM: BAR0 is 0M @ 0x0 (PCI:0000:04:00.0) [97.539919] NVRM: The system BIOS may have misconfigured your GPU. [97.539921] nvidia: probe of 0000:04:00.0 failed with error -1 [97.539966] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid: NVRM: BAR0 is 0M @ 0x0 (PCI:0000:05:00.0) [97.539966] NVRM: The system BIOS may have misconfigured your GPU. [97.539968] nvidia: probe of 0000:05:00.0 failed with error -1 [97.540012] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid: NVRM: BAR0 is 0M @ 0x0 (PCI:0000:06:00.0) [97.540013] NVRM: The system BIOS may have misconfigured your GPU. [97.540015] nvidia: probe of 0000:06:00.0 failed with error -1 [97.540033] NVRM: The NVIDIA probe routine failed for 4 device(s). [97.540034] NVRM: None of the NVIDIA devices were initialized. [97.540154] nvidia-nvlink: Unregistered the Nvlink Core, major device number 238 Anyone have any idea on how I can get this to work? From what I can tell maybe 4 is using too many resources for this platform/motherboard, but I’d like to get some feedback here to see if maybe I can use all 4. ---
**Author:** ![ted4q](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@ted4q](https://forums.developer.nvidia.com/u/ted4q)\ **Post date:** [March 22, 2020, 8:50am UTC](https://forums.developer.nvidia.com/t/cannot-install-tesla-k40-driver-on-ubuntu-18-04/116386/2 "2020-03-22T08:50:44Z")
Forgot to mention in OP, but I am running Ubuntu Server 18.04 and it has been fully updated. ---
**Author:** ![generix](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/generix/32/16024_2.png) [@generix](https://forums.developer.nvidia.com/u/generix)\ **Post date:** [March 25, 2020, 8:34pm UTC](https://forums.developer.nvidia.com/t/cannot-install-tesla-k40-driver-on-ubuntu-18-04/116386/3 "2020-03-25T20:34:26Z")
Not a driver problem, the bios doesn’t provide (enough) 64bit address-space for the Teslas. Please check the bios for an option like ‘above 4G decoding’ or ‘large/64bit BARs’ and enable it. Actually, I don’t believe that a gaming board bios provides that much adress space, mostly server board bioses are designed to achieve it.