Hi,
I have a Ubuntu 16.04 system and installed CUDA 8.0 and driver 375.82 for 2 GTX 1080 Ti card.
±----------------------------------------------------------------------------+
$ lspci|grep VGA
It seems the system can detect both GPU cards, but the driver can only detect one. Any ideas to help ? Thanks!
#Nothing show for “dmesg|grep NVRM”, nothing too with sudo
$ nvidia-smi
±----------------------------------------------------------------------------+
that dmesg output doesn’t make sense to me, since the NVIDIA Driver puts a message in the system log when it is loading, like this:
$ dmesg |grep NVRM
[ 6.655273] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 375.66 Mon May 1 15:29:16 PDT 2017 (using threaded interrupts)
Something there doesn’t add up
maybe you haven’t removed nouveau properly
instead of 375.82 driver, you may want to try 384.59:
http://www.nvidia.com/download/driverResults.aspx/120917/en-us
Make sure to follow the instructions in the CUDA 8 linux install guide concerning removal of nouveau:
http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#abstract
(probably a good idea to read the whole install guide, if you haven’t)
I reinstall the cuda-8.0 without install the default driver and install the driver 384.59.Now “nvidia-smi” also detects only one GPU card.
~$ nvidia-smi
Sun Aug 6 14:05:10 2017
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 384.59 Driver Version: 384.59 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX 108... Off | 00000000:01:00.0 On | N/A |
| 0% 33C P8 11W / 360W | 126MiB / 11169MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 1033 G /usr/lib/xorg/Xorg 74MiB |
| 0 1772 G compiz 41MiB |
| 0 2251 G fcitx-qimpanel 8MiB |
+-----------------------------------------------------------------------------+
~$ dmesg|grep NVRM
[ 2.081639] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 384.59 Wed Jul 19 23:53:34 PDT 2017 (using threaded interrupts)
[ 3.226094] NVRM: Your system is not currently configured to drive a VGA console
[ 4.183854] NVRM: GPU at PCI:0000:02:00: GPU-a29e3dde-b03b-5484-803c-c5bf5b3df99c
[ 4.183857] NVRM: GPU Board Serial Number:
[ 4.183858] NVRM: Xid (PCI:0000:02:00): 62, 1bad(b2f4) 00000000 00000000
[ 62.305789] NVRM: RmInitAdapter failed! (0x26:0xffff:1102)
[ 62.305834] NVRM: rm_init_adapter failed for device bearing minor number 1
[ 75.165485] NVRM: RmInitAdapter failed! (0x26:0xffff:1102)
[ 75.165517] NVRM: rm_init_adapter failed for device bearing minor number 1
Other informations
~$ cat /proc/driver/nvidia/gpus/0000\:01\:00.0/information
Model: GeForce GTX 1080 Ti
IRQ: 135
GPU UUID: GPU-637bcd08-b214-bd17-8b99-035cfea0b6a7
Video BIOS: 86.02.39.00.9c
Bus Type: PCIe
DMA Size: 47 bits
DMA Mask: 0x7fffffffffff
Bus Location: 0000:01:00.0
Device Minor: 0
~$ cat /proc/driver/nvidia/gpus/0000\:02\:00.0/information
Model: GeForce GTX 1080 Ti
IRQ: 136
GPU UUID: GPU-????????-????-????-????-????????????
Video BIOS: ??.??.??.??.??
Bus Type: PCIe
DMA Size: 47 bits
DMA Mask: 0x7fffffffffff
Bus Location: 0000:02:00.0
Device Minor: 1
The loaded nvidia modules:
~$ lsmod|grep nvidia
nvidia_drm 49152 1
nvidia_modeset 843776 5 nvidia_drm
nvidia 13041664 94 nvidia_modeset
drm_kms_helper 155648 2 i915_bpo,nvidia_drm
drm 364544 5 i915_bpo,drm_kms_helper,nvidia_drm
Systems:
~$ uname -a
Linux nisp-dmi-02 4.4.0-89-generic #112-Ubuntu SMP Mon Jul 31 19:38:41 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux
Module information:
~$ modinfo nvidia
filename: /lib/modules/4.4.0-89-generic/kernel/drivers/video/nvidia.ko
alias: char-major-195-*
version: 384.59
supported: external
license: NVIDIA
srcversion: 31FF0349D3C7B1D9A62B474
alias: pci:v000010DEd00000E00sv*sd*bc04sc80i00*
alias: pci:v000010DEd*sv*sd*bc03sc02i00*
alias: pci:v000010DEd*sv*sd*bc03sc00i00*
depends:
vermagic: 4.4.0-89-generic SMP mod_unload modversions
parm: NVreg_Mobile:int
parm: NVreg_ResmanDebugLevel:int
parm: NVreg_RmLogonRC:int
parm: NVreg_ModifyDeviceFiles:int
parm: NVreg_DeviceFileUID:int
parm: NVreg_DeviceFileGID:int
parm: NVreg_DeviceFileMode:int
parm: NVreg_UpdateMemoryTypes:int
parm: NVreg_InitializeSystemMemoryAllocations:int
parm: NVreg_UsePageAttributeTable:int
parm: NVreg_MapRegistersEarly:int
parm: NVreg_RegisterForACPIEvents:int
parm: NVreg_CheckPCIConfigSpace:int
parm: NVreg_EnablePCIeGen3:int
parm: NVreg_EnableMSI:int
parm: NVreg_TCEBypassMode:int
parm: NVreg_UseThreadedInterrupts:int
parm: NVreg_EnableStreamMemOPs:int
parm: NVreg_MemoryPoolSize:int
parm: NVreg_RegistryDwords:charp
parm: NVreg_RegistryDwordsPerDevice:charp
parm: NVreg_RmMsg:charp
parm: NVreg_AssignGpus:charp
The nouveau driver is blacklisted and unloaded:
~$ cat /etc/modprobe.d/blacklist-nouveau.conf
blacklist nouveau
options nouveau modeset=0
Maybe you have insufficient power delivered to the 2nd GPU
I switched the power wire of GPU1 and GPU2, the same issue occured,
herber
February 5, 2018, 2:39am
10
I have encountered these errors over the years and alway struggled to figure out what is really wrong.
I have three 450’s and a 750 in three different computers.
On the machine with a 450 and a 750 the 450 is working great but the 750
[ 62.305789] NVRM: RmInitAdapter failed! (0x26:0xffff:1102)
And this when running nvidia-smi:
GPU UUID: GPU-???-???-???-???-???
The 450’s work any any of my three gentoo computers but the 750 never works.