Dennis HilkLinux Fix LabDE

Graphics, GPU & Display

Application runs on the unintended GPU

Hybrid graphics separates display ownership from rendering; compare device selection per API before forcing the entire desktop onto another GPU.

Editorial source review: Start with the evidenceShare this guide
On this page
  1. Symptoms & scope
  2. Possible causes
  3. Diagnose safely
  4. Evidence-guided next steps
  5. References & review
  6. Related problems

Symptoms & scope

  • An application uses the integrated GPU despite an available discrete GPU.
  • A monitor attached to one GPU works while a rendering application chooses another.

Relevant environment

Multi-GPU systems using Mesa PRIME or NVIDIA render offload; offload support depends on the installed driver and session.

Possible causes

These are possible explanations, not a confirmed diagnosis. Several independent faults can coexist.

  • Default enumeration order or an application’s selection policy may prefer the integrated GPU.
  • OpenGL and Vulkan may use different offload controls; display connector ownership does not establish the rendering GPU.

Diagnose safely

Run one command at a time in the relevant session. Read the explanation first. Uppercase placeholders need your own values; tools and privileges vary by distribution. These commands are displayed here and never executed by the website.

Check 1

List all display controllers and active drivers; no administrative changes are made.

lspci -nnk

Interpret the result: Both GPUs need a suitable active driver for useful offload. An unbound GPU cannot be selected merely by setting an environment variable.

Check 2

Read Vulkan device names in the application’s runtime; requires vulkaninfo.

vulkaninfo --summary

Interpret the result: Multiple devices make per-app selection possible, but enumeration order alone does not prove which device the application actually chose.

Evidence-guided next steps

Select a Mesa GPU for one application

If both Mesa-supported GPUs are listed, test the application with the documented DRI_PRIME selector in its launcher. Confirm the application’s own renderer; Vulkan can require the selector form that hides other devices.

Precautions: Apply the setting per application, not globally. A selected GPU still needs runtime and device access; avoid changing PCI binding.

Recovery / rollback: Remove the added DRI_PRIME setting from the launcher to restore normal device selection.

Did this solution help you?

Share this solution#

Use NVIDIA’s supported render offload

If NVIDIA is the intended rendering GPU, use its documented PRIME offload variables for the specific API: OpenGL GLX also needs NVIDIA vendor selection, while Vulkan uses the NVIDIA offload layer. Check the installed release’s guidance.

Precautions: Do not assume Xorg examples fully configure a Wayland session. Forcing the whole desktop to a discrete GPU can affect battery life and external displays.

Recovery / rollback: Remove only the added offload variables and return to the original application profile.

Did this solution help you?

Share this solution#

References & review

This guide was prepared from primary project or distribution sources and reviewed on the date shown. This is an editorial source check, not evidence that a fix was reproduced on your hardware. Diagnostic log examples are synthetic fixtures. Version-dependent details must be checked against your installed release.