A40 can't change display mode to graphic mode

Graphics / Linux Linux
1

Hi, I try to use displaymodeselector to change to graphic mode for my nvidia A40, but they show :
WARNING: This operation updates the firmware on the board and could make
the device unusable if your host system lacks the necessary support.

Specifed GPU Mode “physical_display_enabled_256MB_bar1”

Device:00:01:00:00=10DE:2235:10DE:145A GPU
ERROR: In order to avoid the irreparable damage to your graphics
adapter it is necessary to unload the NVIDIA kernel driver first:

I tried to : rmmod nvidia_uvm nvidia_drm nvidia_modeset nvidia

They show:
Specifed GPU Mode “physical_display_enabled_256MB_bar1”

Apply GPU Mode <5> corresponds to “physical_display_enabled_256MB_bar1”

Adapter not accessible or supported EEPROM not found, skipping

Reading VBIOS image from EEPROM.

ROM image is not present.

Cannot set GPU mode for this adapter

Nvflash CPU side error Code:2Error Message: Falcon In HALT or STOP state, abort uCode command issuing process.

And nvidia-smi not work after that: No devices were found. Before use displayselector, my A40 always load in ~100W/300 (nothing used) and high temp (90C)

Thank you so much.

2

The A40 doesn’t have fans of its own, it needs the server chassis to provide that. You can’t use it in a regular desktop/workstation enclosures without providing the needed cooling, e.g. printing a funnel on a 3D printer and attaching a fan.

3

thank for your reply. I just put it in server room ~18C. but can’t use the displaymodeselector to change mode. always error like that.

4

Please run nvidia-bug-report.sh as root and attach the resulting nvidia-bug-report.log.gz file to your post.

5

Sometimes i can’t run nvidia-smi : Error: Unable to determine the device handle for GPU 0000:01:00.0: Unknown Error

Try:
sudo nvidia-smi drain -p 0000:01:00.0 -m 0
=>
Failed to modify the GPU drain state.

When I reboot, it’s worked, but usually have this problem. Thank you!

log:
nvidia-bug-report.log (502.5 KB)

6

Again,
The A40 doesn’t have fans of its own,
you can’t run it in a normal desktop, it’s overheating and shutting down. Putting the box in another room doesn’t help.

7

Hi, I don’t use like normal desktop, just remote into for training, I check temperature idle ~30C and 70-80C when training in the server room with high speed fan. It’s good this temp I think, my A6000 have fan and ~ 70C when training.