After completing
[the configuration steps](https://docs.cloud.google.com/confidential-computing/confidential-vm/docs/create-a-confidential-vm-instance-with-gpu),
verify that your GPU is correctly operating in confidential computing mode:

1. Verify that the `nvidia-persistenced` service is running with
   `uvm-persistence-mode` enabled:

       ps aux | grep nvidia-persistenced | grep -v grep

   The output is similar to the following:

       nvidia-+     729  6.2  0.0   5468  1664 ?  Ss   12:02   0:10
       /usr/bin/nvidia-persistenced \
       --user nvidia-persistenced \
       --uvm-persistence-mode \
       --verbose

   > [!CAUTION]
   > **Caution:** Don't restart the `nvidia-persistenced` service after enabling `uvm-persistence-mode`, as this can disrupt the secure connection and require a VM reboot.

2. Confirm that the GPU is running in confidential computing mode:

       sudo nvidia-smi conf-compute -f

   The output is similar to the following:

       CC status: ON

3. Set GPU to ready state after each reboot.

       sudo nvidia-smi conf-compute -srs 1

4. Verify Confidential Computing ready state.

       sudo nvidia-smi conf-compute -grs

   If you transition a GPU out of the ready state using
   `nvidia-smi conf-compute -srs 0`, reboot the VM instance to return the GPU
   to the ready state.

   The output is similar to the following:

       Confidential Compute GPUs Ready state: ready