Windows Server 2019 + NVIDIA GeForce GTX 750 Ti = cuInit(): no CUDA-capable device
#1
Hello!

hashcat don't see my NVIDIA GeForce GTX 750 Ti: "cuInit(): no CUDA-capable device is detected"
Please ask how to fix it.
Thank you.

Code:
Windows Server 2019 1808 (10.0.17763.4974)
NVIDIA GeForce GTX 750 Ti
545.84
DCH
video BIOS 82.07.32.40.47

nvui.dll 8.17.15.4584 NVIDIA User Experience Driver Component
nvxdplcy.dll 8.17.15.4584 NVIDIA User Experience Driver Component
nvxdbat.dll 8.17.15.4584 NVIDIA User Experience Driver Component
nvxdapix.dll 8.17.15.4584 NVIDIA User Experience Driver Component
NVCPL.DLL 8.17.15.4584 NVIDIA User Experience Driver Component
nvCplUIR.dll 8.1.940.0 NVIDIA Control Panel
nvCplUI.exe 8.1.940.0 NVIDIA Control Panel
nvWSSR.dll 31.0.15.4584 NVIDIA Workstation Server
nvWSS.dll 31.0.15.4584 NVIDIA Workstation Server
nvDevToolSR.dll 31.0.15.4584 NVIDIA Licensing Server
nvDevToolS.dll 31.0.15.4584 NVIDIA 3D Settings Server
NVCUDA64.DLL 31.0.15.4584 NVIDIA CUDA 12.3.68 driver
nvGameSR.dll 31.0.15.4584 NVIDIA 3D Settings Server
nvGameS.dll 31.0.15.4584 NVIDIA 3D Settings Server
nvLicensingS.dll 6.14.15.4584 NVIDIA Licensing Server
nvDispSR.dll 31.0.15.4584 NVIDIA Display Server
nvDispS.dll 31.0.15.4584 NVIDIA Display Server
nvViTvSR.dll 31.0.15.4584 NVIDIA Video Server
nvViTvS.dll 31.0.15.4584 NVIDIA Video Server

Code:
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.3\extras\demo_suite\deviceQuery.exe
Starting...

CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "NVIDIA GeForce GTX 750 Ti"
  CUDA Driver Version / Runtime Version          12.3 / 12.3
  CUDA Capability Major/Minor version number:    5.0
  Total amount of global memory:                2048 MBytes (2147221504 bytes)
  ( 5) Multiprocessors, (128) CUDA Cores/MP:    640 CUDA Cores
  GPU Max Clock rate:                            1202 MHz (1.20 GHz)
  Memory Clock rate:                            2700 Mhz
  Memory Bus Width:                              128-bit
  L2 Cache Size:                                2097152 bytes
  Maximum Texture Dimension Size (x,y,z)        1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
  Maximum Layered 1D Texture Size, (num) layers  1D=(16384), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(16384, 16384), 2048 layers
  Total amount of constant memory:              zu bytes
  Total amount of shared memory per block:      zu bytes
  Total number of registers available per block: 65536
  Warp size:                                    32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:          1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          zu bytes
  Texture alignment:                            zu bytes
  Concurrent copy and kernel execution:          Yes with 1 copy engine(s)
  Run time limit on kernels:                    Yes
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:      Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  CUDA Device Driver Mode (TCC or WDDM):        WDDM (Windows Display Driver Model)
  Device supports Unified Addressing (UVA):      Yes
  Device supports Compute Preemption:            No
  Supports Cooperative Kernel Launch:            No
  Supports MultiDevice Co-op Kernel Launch:      No
  Device PCI Domain ID / Bus ID / location ID:  0 / 8 / 0
  Compute Mode:
    < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 12.3, CUDA Runtime Version = 12.3, NumDevs = 1, Device0 = NVIDIA GeForce GTX 750 Ti
Result = PASS



Code:
hashcat.exe -I 
hashcat (v6.2.6) starting in backend information mode

cuInit(): no CUDA-capable device is detected

OpenCL Info:
============

OpenCL Platform ID #1
  Vendor..: Intel(R) Corporation
  Name....: Intel(R) OpenCL
  Version.: OpenCL 2.1 WINDOWS

  Backend Device ID #1
    Type...........: CPU
    Vendor.ID......: 8
    Vendor.........: Intel(R) Corporation
    Name...........: AMD Ryzen 7 2700X Eight-Core Processor
    Version........: OpenCL 2.1 (Build 0)
    Processor(s)...: 16
    Clock..........: 0
    Memory.Total...: 65458 MB (limited to 8182 MB allocatable in one block)
    Memory.Free....: 32697 MB
    Local.Memory...: 32 KB
    OpenCL.Version.: OpenCL C 2.0
    Driver.Version.: 2021.11.3.0.17_160000


Code:
hashcat.exe -I
hashcat (v6.2.6-747-g5b06ffe63) starting in backend information mode

Support for HIPRTC was dropped by AMD Adrenalin Edition 22.7.1 and later.
This is not a hashcat problem.

Please install the AMD HIP SDK

cuInit(): no CUDA-capable device is detected

OpenCL Info:
============

OpenCL Platform ID #1
  Vendor..: Intel(R) Corporation
  Name....: Intel(R) OpenCL
  Version.: OpenCL 2.1 WINDOWS

  Backend Device ID #1
    Type...........: CPU
    Vendor.ID......: 8
    Vendor.........: Intel(R) Corporation
    Name...........: AMD Ryzen 7 2700X Eight-Core Processor
    Version........: OpenCL 2.1 (Build 0)
    Processor(s)...: 16
    Clock..........: 0
    Memory.Total...: 65458 MB (limited to 8182 MB allocatable in one block)
    Memory.Free....: 32697 MB
    Local.Memory...: 32 KB
    OpenCL.Version.: OpenCL C 2.0
    Driver.Version.: 2021.11.3.0.17_160000
Reply
#2
Please double-check if NVidia do officially support the specific operating system you want to use.

Read here
https://hashcat.net/wiki/doku.php?id=fre..._the_gpu_s
https://hashcat.net/faq/wrongdriver

Try a freshly installed Windows 10 (not Windows Server).
Also you can try to isolate the problem using Linux.
Please see here
https://hashcat.net/forum/thread-11349.html
https://wiki.archlinux.org/title/GPGPU
Reply
#3
I believe this is not a hardware problem.
And I also believe that this hardware will work with hashcat on Linux or Windows 10.
But this PC is not a "hashcat PC", I can only run hashcat at a time when it is free from other tasks.
And deviceQuery.exe from NVIDIA sees the video card, so I assume that drivers are correct.

BTW. Now I use for hashcat OpenCL 2.1 with AMD Ryzen 7 2700X on this PC. CPU load is 100% all the time while hashcat is running. Can I reduce the CPU load? I tried "-w 1", but nothing changed.
Reply
#4
If possible, try to boot a fresh Windows (or Linux) from a USB-flash.
Then install the correct drivers and play with hashcat there.
Otherwise, it is difficult to accurately diagnose the problem.

Ubuntu is recommended Linux distribution.
https://hashcat.net/forum/thread-8026-po...l#pid43147
https://hashcat.net/forum/thread-9587-po...l#pid50503
https://hashcat.net/forum/thread-6661-po...l#pid40381

Concerning CPU load, may be these posts will be useful
https://hashcat.net/forum/thread-5184.html
https://hashcat.net/forum/thread-2999.html
Reply
#5
Maybe anybody has any ideas how to use NVIDIA GeForce GTX 750 Ti in hashcat on Windows Server 2019?
(I can't change OS.)
Thank you.
Reply
#6
Which drivers did you install for your graphics card? It is not detecting an usable driver in either the format of OpenCL (Nvidia Graphics Drivers) or CUDA (CUDA Developer Toolkit). Judging by your setup, it has several drivers installed but were those installed by yourself or system admin? Preloaded by windows? What authorization do you have on the PC? Full admin privileges? User privileges? Are you running in a VM?

Otherwise, once hashcat detects a proper runtime it can use, it will be defined in hashcat -I and you will need to isolate the device to only using the GPU by either using -d or -D, which ever is more convenient for yourself.

This all could be in relation to enterprise drivers not being compatible but that would require a confirmation from someone else using the operating system or maybe a developer can chirp in.
Reply
#7
slyexe, thanks for your response!

This is a physical PC without virtualization, I connect via RDP, I am the only administrator with full rights.
The drivers were originally installed with Windows by the vendor.
Then I downloaded cuda_12.3.0_545.84_windows.exe from the Nvidia website.
When installing cuda_12.3.0_545.84_windows.exe, it was written that my old driver (I don’t remember the exact version) would be replaced with the driver version 545.84 available in this package.

Based on the fact that
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.3\extras\demo_suite\deviceQuery.exe
sees CUDA (listing in the first post), I'm guessing this is some kind of hashcat issue. So I wanted to get advice from more knowledgeable forum members or provide additional information for a full bug report.
Reply
#8
Has anybody any new ideas maybe?
Reply
#9
Please see below How to set minimal workload
https://github.com/hashcat/hashcat/issues/2781
Reply
#10
Don't assume your driver is correct, reinstall as stated here
https://hashcat.net/faq/wrongdriver

Do you have physical access to the server?
Reply