For multiple GPUs: torch.cuda.empty_cache() stuck forever

[animeshkumarpaul]

System Info

• transformers version: 4.41.0.dev0
• Platform: Linux-5.4.0-172-generic-x86_64-with-glibc2.31
• Python version: 3.11.0
• Huggingface_hub version: 0.23.0
• Safetensors version: 0.4.3
• Accelerate version: 0.30.0
• Accelerate config: not found
• PyTorch version (GPU?): 2.2.2 (True)
• Tensorflow version (GPU?): not installed (NA)
• Flax version (CPU?/GPU?/TPU?): not installed (NA)
• Jax version: not installed
• JaxLib version: not installed
• Using GPU in script?: yes
• Using distributed or parallel set-up in script?: yes

Who can help?

@muellerzr and @pacman10

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
• My own task or dataset (give details below)

Reproduction

transformers/src/transformers/trainer.py

Line 3219 in 835de4c

torch.cuda.empty_cache()

I am using the BartForConditionalGeneration model.

Expected behavior

For the multiple GPUs: during the training time, the process stucks at this line forever - at that time there is no GPU usage, but there is CPU usage.