Cant passthrough laptop nvidia gpu in windows 11

Open
#86 1 comment 3 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
35/100
Issue type
Bug
Clarity
Needs clarification
Activity status
Active

Research direction

Start by reproducing the GPU-PV setup on Windows Canary 26200 with the NVIDIA RTX 3050 and driver 610.62, then inspect the AppSandbox source's active-driver queries and GPU-PV device initialization. Compare Device Manager, DriverStore, pnputil, sc query nvlddmkm, driverquery, and nvidia-smi results to determine whether the driver injection or virtual hardware ID binding fails.

Written by the indexing model from the issue text.

Description

My laptop integrated intel gpu works fine but when i passthrough by using TryAll , it shows nvidia gpu in task manager,device manager but not showing in any other places like during nvidia driver installation , gpu-test etc.

I'm using the latest AppSandbox on Windows Insider Canary 26200.8737 with an NVIDIA RTX 3050 6GB Laptop GPU and NVIDIA driver 610.62 (CUDA 13.3). On the host, everything works correctly: nvidia-smi works, but inside the Sandbox GPU acceleration does not work. Initially, Device Manager showed the NVIDIA GPU with Code 43, alongside Intel UHD Graphics, App Sandbox Virtual Display Adapter, and Hyper-V Video. The guest DriverStore contains nv_dispig.inf_*, while the host uses oem98.inf (original INF nvhmsi.inf), so I investigated whether the wrong driver package was being injected. After manually extracting and installing the NVIDIA display driver inside the Sandbox, Device Manager now reports "This device is working properly", but the driver provider is still Microsoft, not NVIDIA. There is no nvlddmkm service (sc query nvlddmkm and the corresponding registry key return nothing), driverquery shows no NVIDIA display driver, nvidia-smi is not present, and NVIDIA App reports "Requires an NVIDIA GPU" even though nvcuda.dll exists. Attempts to install the extracted driver using pnputil fail (e.g., "missing or invalid driver package specified"), and the NVIDIA kernel driver never binds to the virtual device. I inspected the current AppSandbox source and did not find any obvious hardcoded assumptions about old nv_disp*.inf layouts—the code appears to query the active driver dynamically—so this does not appear to be a simple driver-copy issue. Could this be a regression or incompatibility between Windows Canary 26200, NVIDIA 610.62, and AppSandbox's GPU-PV device initialization, or is the NVIDIA driver failing to bind because the GPU-PV virtual hardware ID is not being matched correctly? Any guidance on debugging or validating the GPU driver injection/binding path would be greatly appreciated.

Dominant language
C
Stars
704
Forks
75
Avg merge
9m
Merged PRs (30d)
6

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from jamesstringer90/appsandbox

All issues in jamesstringer90/appsandbox

Similar issues

More C issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.