[model][refactor] remove cuda hard code in models and layers

Signed-off-by: Mengqing Cao <cmq0113@163.com>

Author: MengqingCao

vllm/model_executor/layers/fused_moe/fused_marlin_moe.py

@@ -7,6 +7,7 @@
 
 from vllm.model_executor.layers.fused_moe.fused_moe import (
     fused_topk, moe_align_block_size, try_get_optimal_moe_config)
+from vllm.platforms import current_platform
 from vllm.scalar_type import scalar_types
 from vllm.utils import direct_register_custom_op
 
@@ -238,7 +239,7 @@ def fused_marlin_moe(
     max_workspace_size = (max(2 * N, K) // 64) * 16
     workspace = torch.zeros(max_workspace_size,
                             dtype=torch.int,
-                            device="cuda",
+                            device=current_platform.device_type,
                             requires_grad=False)
 
     if has_no_zp:

vllm/model_executor/layers/rotary_embedding.py

@@ -30,6 +30,7 @@
 from transformers import PretrainedConfig
 
 from vllm.model_executor.custom_op import CustomOp
+from vllm.platforms import current_platform
 
 
 def _rotate_neox(x: torch.Tensor) -> torch.Tensor:
@@ -651,7 +652,7 @@ def __init__(
 
     def _compute_inv_freq(self, scaling_factor: float) -> torch.Tensor:
         pos_freqs = self.base**(torch.arange(
-            0, self.rotary_dim, 2, dtype=torch.float, device="cuda") /
+            0, self.rotary_dim, 2, dtype=torch.float, device=current_platform.device_type) /
                                 self.rotary_dim)
         inv_freq_extrapolation = 1.0 / pos_freqs
         inv_freq_interpolation = 1.0 / (scaling_factor * pos_freqs)
@@ -670,7 +671,7 @@ def _compute_inv_freq(self, scaling_factor: float) -> torch.Tensor:
     def _compute_cos_sin_cache(self) -> torch.Tensor:
         inv_freq = self._compute_inv_freq(self.scaling_factor)
         t = torch.arange(self.max_position_embeddings * self.scaling_factor,
-                         device="cuda",
+                         device=current_platform.device_type,
                          dtype=torch.float32)
         freqs = torch.einsum("i,j -> ij", t, inv_freq)
         cos = (freqs.cos() * self.mscale)

vllm/model_executor/layers/spec_decode_base_sampler.py

@@ -6,7 +6,7 @@
 import torch
 import torch.jit
 import torch.nn as nn
-
+from vllm.platforms import current_platform
 
 class SpecDecodeBaseSampler(nn.Module):
     """Base class for samplers used for Speculative Decoding verification
@@ -35,7 +35,7 @@ def __init__(self, strict_mode: bool = False):
     def init_gpu_tensors(self, device: Union[int, str]) -> None:
         assert self.num_accepted_tokens is None
         if isinstance(device, int):
-            device = f"cuda:{device}"
+            device = f"{current_platform.device_type}:{device}"
         elif not isinstance(device, str):
             raise ValueError(f"Device must be int or str, get {type(device)}")
         self.num_accepted_tokens = torch.tensor(0,

vllm/model_executor/model_loader/loader.py

@@ -914,7 +914,7 @@ def _parse_quant_state(param_name: str,
                 if param_name + "." in k:
                     quant_state[k] = temp_state_dict[k]
 
-            return QuantState.from_dict(quant_state, device="cuda")
+            return QuantState.from_dict(quant_state, device=current_platform.device_type)
 
         # Second iterate over all prequant and normal weights
         # pre quantized weights would have a quant_state

vllm/model_executor/models/arctic.py

@@ -30,6 +30,7 @@
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 from vllm.model_executor.sampling_metadata import SamplingMetadata
 from vllm.model_executor.utils import set_weight_attrs
+from vllm.platforms import current_platform
 from vllm.sequence import IntermediateTensors
 from vllm.transformers_utils.configs.arctic import ArcticConfig
 
@@ -138,13 +139,13 @@ def __init__(self,
                     torch.empty(self.num_experts,
                                 2 * self.intermediate_size,
                                 self.hidden_size,
-                                device="cuda",
+                                device=current_platform.device_type,
                                 dtype=self.params_dtype))
                 self.w2s = nn.Parameter(
                     torch.empty(self.num_experts,
                                 self.hidden_size,
                                 self.intermediate_size,
-                                device="cuda",
+                                device=current_platform.device_type,
                                 dtype=self.params_dtype))
             set_weight_attrs(self.ws, {
                 "weight_loader": self.weight_loader,

vllm/model_executor/models/minicpm.py

@@ -51,6 +51,7 @@
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 from vllm.model_executor.sampling_metadata import SamplingMetadata
 from vllm.model_executor.utils import set_weight_attrs
+from vllm.platforms import current_platform
 from vllm.sequence import IntermediateTensors
 
 from .interfaces import SupportsLoRA, SupportsPP
@@ -98,13 +99,13 @@ def __init__(
             torch.empty(self.num_total_experts,
                         2 * self.intermediate_size,
                         self.hidden_size,
-                        device="cuda",
+                        device=current_platform.device_type,
                         dtype=self.params_dtype))
         self.w2s = nn.Parameter(
             torch.empty(self.num_total_experts,
                         self.hidden_size,
                         self.intermediate_size,
-                        device="cuda",
+                        device=current_platform.device,
                         dtype=self.params_dtype))
 
         set_weight_attrs(self.ws, {

vllm/model_executor/models/minicpmv.py

@@ -59,6 +59,7 @@
 from vllm.multimodal.processing import (BaseMultiModalProcessor,
                                         BaseProcessingInfo, PromptReplacement)
 from vllm.multimodal.profiling import BaseDummyInputsBuilder, ProcessorInputs
+from vllm.platforms import current_platform
 from vllm.sequence import IntermediateTensors
 
 from .idefics2_vision_model import Idefics2VisionTransformer
@@ -1184,7 +1185,7 @@ def init_resampler(self,
                                    quant_config=quant_config,
                                    prefix=prefix)
 
-        return resampler.to(device="cuda", dtype=torch.get_default_dtype())
+        return resampler.to(device=current_platform.device_type, dtype=torch.get_default_dtype())
 
     def get_vision_embedding(
         self,
@@ -1266,7 +1267,7 @@ def init_resampler(self,
                                      quant_config=quant_config,
                                      prefix=prefix)
 
-        return resampler.to(device="cuda", dtype=torch.get_default_dtype())
+        return resampler.to(device=current_platform.device_type, dtype=torch.get_default_dtype())
 
     def get_vision_embedding(
         self,
@@ -1360,7 +1361,7 @@ def init_resampler(self,
                                      quant_config=quant_config,
                                      prefix=prefix)
 
-        return resampler.to(device="cuda", dtype=torch.get_default_dtype())
+        return resampler.to(device=current_platform.device_type, dtype=torch.get_default_dtype())
 
     def get_vision_embedding(
         self,