Cannot Disable ECC on L40 on Ubuntu 24.04

[leroy8]

all,

Not sure if this is the best place to post this.

I have a Linux Ubuntu server running v24.04 with a L4 and an L40 GPU installed in it.
Running Nvidia driver version 555.42.06

When i run the command “nvidia-smi -e 0” it disables ECC on both GPU’s and that is good.
It then says i need to reboot. So i do.

When the system comes back up… the L4 has ECC disabled but the L40 does not.
So I try again on only the L40 with “nvidia-smi -i 00000000:CA:00.0 -e 0”… then reboot.
Same thing… ECC still enabled. No matter what I try I cannot disable ECC on the L40.

Any suggestions?

here is the output from the nividia-smi tool

Thu Jul 25 13:25:22 2024
±----------------------------------------------------------------------------------------+
| NVIDIA-SMI 555.42.06 Driver Version: 555.42.06 CUDA Version: 12.5 |
|-----------------------------------------±-----------------------±---------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA L4 Off | 00000000:65:00.0 Off | Off |
| N/A 66C P0 70W / 72W | 24125MiB / 24570MiB | 100% Default |
| | | N/A |
±----------------------------------------±-----------------------±---------------------+
| 1 NVIDIA L40 Off | 00000000:CA:00.0 Off | 0 |
| N/A 65C P0 301W / 300W | 24749MiB / 46068MiB | 100% Default |
| | | N/A |
±----------------------------------------±-----------------------±---------------------+

Thanks

[leroy8]

I also tried with Ubuntu 22.04 and with the 535 version of the driver… same results.

[leroy8]

I tried using the version 225 of the “server” flavor of the driver on Ubuntu 24.04 and this worked.

Don’t know if it was the version change or the server type of the driver… but this is solved for me now.