hollowaykeanho / Upscaler

A consolidation of various compiled open-source AI image/video upscaling product for a working CLI friendly image and video upscaling program.
BSD 3-Clause "New" or "Revised" License
252 stars 20 forks source link

BUG: vkCreateInstance failed -9 invalid gpu device #33

Open ihorizons2022 opened 1 year ago

ihorizons2022 commented 1 year ago
image

used in colab, GPU: tesla t4

hollowaykeanho commented 1 year ago

Thanks. Acknowledged.

Tracking this issue: https://github.com/Tencent/ncnn/issues/1548

Also, can you publish the list of GPU (printed before any frames) please? I will add the GPU controller in.

Also, also: $ nvidia-smi

hollowaykeanho commented 1 year ago

Also on Ubuntu, is your apt install mesa-vulkan-drivers done?

Tracking: https://github.com/nihui/realsr-ncnn-vulkan/issues/18


更新:您可以运用华语。

ihorizons2022 commented 1 year ago

thanks for your reply, i will have a try and return the result back

ihorizons2022 commented 1 year ago

Mon Jul 17 10:59:14 2023
+-----------------------------------------------------------------------------+ | NVIDIA-SMI 525.105.17 Driver Version: 525.105.17 CUDA Version: 12.0 | |-------------------------------+----------------------+----------------------+ | 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 Tesla T4 Off | 00000000:00:04.0 Off | 0 | | N/A 37C P8 9W / 70W | 0MiB / 15360MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | No running processes found | +-----------------------------------------------------------------------------+

i install mesa-vulkan-drivers on ubuntu, but no luck, still cannot work properly

image
hollowaykeanho commented 1 year ago

The GPU output is nvidia-smi, right?

ihorizons2022 commented 1 year ago

yes

ihorizons2022 commented 1 year ago
image
hollowaykeanho commented 1 year ago

No luck. It means the binary does not work with Nvidia Tesla (like how I tested on Apple Iris). You got 2 choices:

1) Use another machines. Known tested one are listed on the benchmark section. 2) Wait for me to complete this ticket - this can take time.

I'll add your test results into the benchmark list. Thanks for the input.

hollowaykeanho commented 1 year ago

Dependent on Issue #34 and the entire project will now entering program re-write phase.

ihorizons2022 commented 1 year ago

thank you very much

hollowaykeanho commented 1 year ago

no problem.