Resolving a graphics processing unit problem

Learn about the possible problems and service actions that you can perform to resolve a graphics processing unit (GPU) problem.

Note: To determine the location of the GPU, see Identifying the location of the GPU by using the slot number.
Table 1. GPU problems and service actions
Problem Service action
System unable to find GPU
  1. Verify that the GPU is properly seated in a compatible slot.
  2. Install the GPU in a different compatible slot.
  3. Verify that the drivers for the GPU are installed.
  4. Verify that the most recent firmware is installed on the system, or install the most recent firmware if it is not already installed.
  5. Restart the system.
  6. Replace the GPU.
  7. Replace the system backplane.
  8. Replace the central processing unit (CPU).
GPU suddenly stops working
  1. If the system was recently installed, moved, serviced, or upgraded, verify that the GPU is seated properly and all associated cables are connected correctly.
  2. Inspect the PCIe socket and verify that there is no dirt or debris in the socket.
  3. Inspect the card and verify that it is not physically damaged.
  4. Verify that all cables are properly seated and are not physically damaged. If you recently added one or more new adapters, remove them and then test to determine whether the failing adapter is functioning properly again. If the graphics adapter is functioning again, review the IBM support tips to confirm that there are no PCI address, driver, or firmware conflicts. Then, reinstall the new adapters again one at a time until all adapters function properly.
  5. Replace the GPU.
  6. Replace the system backplane.
  7. Replace the central processing unit (CPU).
Other problems For information about adapter diagnostics, see Supporting diagnostics. For information about adapter user information, see User guides for GPUs and PCIe adapters.
Parent topic: Resolving a GPU, PCIe adapter, or device problem


Last updated: Thu, December 02, 2021