Re: Segmentation fault for the tutorial

From: Mahmood Naderan (mahmood.nt_at_gmail.com)
Date: Wed Mar 07 2018 - 01:33:55 CST

The OS is Ubuntu and I installed the cuda-9 driver which comes from
the toolkit package. The installation was successful and I can work
with the nvccc, and other things. For example, the Nvidia X server
setting window correctly shows the device, utilization, temperature
and other things.

Your sentence that the driver version must match smi version is not
clear for me. Can you explain more? In what situations they might be
different?

Regards,
Mahmood

On Tue, Mar 6, 2018 at 11:57 PM, Vermaas, Joshua
<Joshua.Vermaas_at_nrel.gov> wrote:
> Hi Mahmood,
>
> I've built it before myself on CUDA, so I know it works for some value
> of compiler/runtime options. The usual things to check when the CPU
> build works but the GPU doesn't are things like the output of nvidia-smi
> on the machine, to make sure that the driver version matches the runtime
> version, and that both versions are recent enough for the CUDA build you
> have built against.
>
> This is what the header looks like for me:
>
> +-----------------------------------------------------------------------------+
> | NVIDIA-SMI 384.111 Driver Version:
> 384.111 |
> |-------------------------------+----------------------+----------------------+
>
> Which confirms that the driver (right) matches the installed runtime
> (left). For me, this driver version would not be enough to run CUDA 9.0,
> since that requires a minimum driver version of 390 (I think).
>
> -Josh

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2018 - 23:20:54 CST