tests: revert change of torch_require_multi_gpu to be device agnostic

src/transformers/integrations/bitsandbytes.py

@@ -486,7 +486,7 @@ def _validate_bnb_multi_backend_availability(raise_exception):
     import bitsandbytes as bnb
 
     bnb_supported_devices = getattr(bnb, "supported_torch_devices", set())
-    available_devices = get_available_devices()
+    available_devices = set(get_available_devices())
 
     if available_devices == {"cpu"} and not is_ipex_available():
         from importlib.util import find_spec

src/transformers/testing_utils.py

@@ -237,17 +237,6 @@ def parse_int_from_env(key, default=None):
 _run_third_party_device_tests = parse_flag_from_env("RUN_THIRD_PARTY_DEVICE_TESTS", default=False)
 
 
-def get_device_count():
-    import torch
-
-    if is_torch_xpu_available():
-        num_devices = torch.xpu.device_count()
-    else:
-        num_devices = torch.cuda.device_count()
-
-    return num_devices
-
-
 def is_staging_test(test_case):
     """
     Decorator marking a test as a staging test.
@@ -755,17 +744,17 @@ def require_spacy(test_case):
 
 def require_torch_multi_gpu(test_case):
     """
-    Decorator marking a test that requires a multi-GPU setup (in PyTorch). These tests are skipped on a machine without
-    multiple GPUs.
+    Decorator marking a test that requires a multi-GPU CUDA setup (in PyTorch). These tests are skipped on a machine without
+    multiple CUDA GPUs.
 
     To run *only* the multi_gpu tests, assuming all test names contain multi_gpu: $ pytest -sv ./tests -k "multi_gpu"
     """
     if not is_torch_available():
         return unittest.skip(reason="test requires PyTorch")(test_case)
 
-    device_count = get_device_count()
+    import torch
 
-    return unittest.skipUnless(device_count > 1, "test requires multiple GPUs")(test_case)
+    return unittest.skipUnless(torch.cuda.device_count() > 1, "test requires multiple CUDA GPUs")(test_case)
 
 
 def require_torch_multi_accelerator(test_case):

tests/quantization/bnb/test_4bit.py

@@ -38,7 +38,7 @@
     require_bitsandbytes,
     require_torch,
     require_torch_gpu_if_bnb_not_multi_backend_enabled,
-    require_torch_multi_gpu,
+    require_torch_multi_accelerator,
     slow,
     torch_device,
 )
@@ -514,7 +514,7 @@ def test_pipeline(self):
         self.assertIn(pipeline_output[0]["generated_text"], self.EXPECTED_OUTPUTS)
 
 
-@require_torch_multi_gpu
+@require_torch_multi_accelerator
 @apply_skip_if_not_implemented
 class Bnb4bitTestMultiGpu(Base4bitTest):
     def setUp(self):

tests/quantization/bnb/test_mixed_int8.py

@@ -39,7 +39,7 @@
     require_bitsandbytes,
     require_torch,
     require_torch_gpu_if_bnb_not_multi_backend_enabled,
-    require_torch_multi_gpu,
+    require_torch_multi_accelerator,
     slow,
     torch_device,
 )
@@ -669,7 +669,7 @@ def test_pipeline(self):
         self.assertIn(pipeline_output[0]["generated_text"], self.EXPECTED_OUTPUTS)
 
 
-@require_torch_multi_gpu
+@require_torch_multi_accelerator
 @apply_skip_if_not_implemented
 class MixedInt8TestMultiGpu(BaseMixedInt8Test):
     def setUp(self):
@@ -698,7 +698,7 @@ def test_multi_gpu_loading(self):
         self.assertIn(self.tokenizer.decode(output_parallel[0], skip_special_tokens=True), self.EXPECTED_OUTPUTS)
 
 
-@require_torch_multi_gpu
+@require_torch_multi_accelerator
 @apply_skip_if_not_implemented
 class MixedInt8TestCpuGpu(BaseMixedInt8Test):
     def setUp(self):