nvidia-smi gpu name error

Hello. I have 2 gtx580 cards. They work fine, but one moment confuse me.
nvidia-smi can’t correctly determine gpu name for second card:
±-----------------------------------------------------+
| NVIDIA-SMI 4.304.54 Driver Version: 304.54 |
|-------------------------------±---------------------±---------------------+
| GPU Name | Bus-Id Disp. | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX 580 | 0000:01:00.0 N/A | N/A |
| 54% 65C N/A N/A / N/A | 43% 665MB / 1535MB | N/A Default |
±------------------------------±---------------------±---------------------+
| 1 ERR! | 0000:07:00.0 N/A | N/A |
| 48% 59C N/A N/A / N/A | 43% 665MB / 1535MB | N/A Default |
±------------------------------±---------------------±---------------------+

What is it?

nvidia-smi is targeted at the Tesla™ and Quadro™ devices, though limited support is also available on other NVIDIA GPUs. Suggest you upgrade to the latest driver 319.23 via http://www.nvidia.com/object/linux-display-amd64-319.23-driver.html. It’s compatible with CUDA 5.0.

If this problem persists, please file a bug report using the bug reporting form linked from the registered developer website. Please run the following

Linux:

__NVML_DBG_FILE=nvml_trace.log __NVML_DBG_LVL=DEBUG nvidia-smi -q &> nvsmi_output.txt

Windows:

set __NVML_DBG_FILE=nvml_trace.log
set __NVML_DBG_LVL=DEBUG
nvidia-smi -q &> nvsmi_output.txt

and provide both nvml_trace.log and nvsmi_output.txt in the bug report. Thank you.

njuffa, sorry, but I cant find “bug reporting form”. Is it feedback button between “Logout” and “My account”?

(1) Login at https://developer.nvidia.com/user/login
(2) Click green link “CUDA/GPU Computing Registered Developer Program”
(3) Under heading “Reporting An Issue” click green link “The Submit a Bug Form”

Thanx. The trouble was that my “CUDA/GPU Computing Registered Developer Program” status was not approved. Now all is ok.