add support for flashcomm2 in qwen3

David9857 · David9857 · commit 8a8d615fba27 · 2025-07-10T19:16:19.000+08:00
Signed-off-by: David9857 &lt;985700846@qq.com&gt;

fix

Signed-off-by: David9857 &lt;985700846@qq.com&gt;

remove self.pad_size

Signed-off-by: David9857 &lt;985700846@qq.com&gt;
diff --git a/vllm_ascend/envs.py b/vllm_ascend/envs.py
@@ -143,6 +143,8 @@
     # Batch MC2 in prefill: The number of tokens in each batch
     "VLLM_ASCEND_FUSED_MOE_MC2_CHUNK_SIZE":
     lambda: int(os.getenv("VLLM_ASCEND_FUSED_MOE_MC2_CHUNK_SIZE", "128")),
+    "VLLM_ENABLE_FC":
+    lambda: int(os.getenv("VLLM_ENABLE_FC", 0))
 }
 
 # end-env-vars-definition
diff --git a/vllm_ascend/models/qwen3.py b/vllm_ascend/models/qwen3.py
@@ -3,37 +3,227 @@
 
 import torch
 from torch import nn
+import torch.nn.functional as F
+import torch.distributed as dist
 from transformers import Qwen3Config
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
-from vllm.distributed import get_pp_group
+from vllm.attention import Attention, AttentionType
+from vllm.distributed import (get_pp_group, 
+                              get_tensor_model_parallel_world_size,
+                              get_tensor_model_parallel_rank,
+                              tensor_model_parallel_all_gather)
 from vllm.model_executor.layers.logits_processor import LogitsProcessor
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.layers.vocab_parallel_embedding import ParallelLMHead
+from vllm.model_executor.layers.linear import RowParallelLinear, ReplicatedLinear
 from vllm.model_executor.models.interfaces import SupportsLoRA, SupportsPP
 from vllm.model_executor.models.qwen2 import Qwen2Model
-from vllm.model_executor.models.qwen3 import Qwen3DecoderLayer
+from vllm.model_executor.models.qwen2 import Qwen2MLP as Qwen3MLP
+from vllm.model_executor.models.qwen3 import Qwen3DecoderLayer, Qwen3Attention
 from vllm.model_executor.models.utils import (AutoWeightsLoader,
                                               PPMissingLayer, maybe_prefix)
 from vllm.model_executor.sampling_metadata import SamplingMetadata
 from vllm.sequence import IntermediateTensors
 
+from vllm_ascend import envs
 from vllm_ascend.ops.layernorm import AddRMSNormQuant
 
 
-class CustomQwen3DecoderLayer(Qwen3DecoderLayer):
+def pad(tensor, x):
+    length = tensor.size(0)
+    pad_size = (x - (length % x)) % x
+    if pad_size > 0:
+        return F.pad(tensor, (0, 0, 0, pad_size)), pad_size
+    return tensor, pad_size
+
+def unpad(tensor, pad_size):
+    if pad_size > 0:
+        return tensor[:-pad_size, :]
+    return tensor
+
+
+class CustomQwen3MLP(Qwen3MLP):
 
     def __init__(
         self,
-        config: Qwen3Config,
-        cache_config: Optional[CacheConfig] = None,
+        hidden_size: int,
+        intermediate_size: int,
+        hidden_act: str,
         quant_config: Optional[QuantizationConfig] = None,
         prefix: str = "",
     ) -> None:
-        super().__init__(config=config,
-                         cache_config=cache_config,
+        super().__init__(hidden_size=hidden_size,
+                         intermediate_size=intermediate_size,
+                         hidden_act=hidden_act,
                          quant_config=quant_config,
                          prefix=prefix)
+        self.tp_size = get_tensor_model_parallel_world_size()
+        self.tp_rank = get_tensor_model_parallel_rank()
+        self.enable_fc = envs.VLLM_ENABLE_FC
+        if self.enable_fc:
+            # if flashcomm2 enbaled, replace Linear+AllReduce with All2All+Linear
+            self.down_proj = ReplicatedLinear(
+                intermediate_size,
+                hidden_size,
+                bias=False,
+                quant_config=quant_config,
+                prefix=f"{prefix}.down_proj",
+            )
+        else:
+            self.down_proj = RowParallelLinear(
+                intermediate_size,
+                hidden_size,
+                bias=False,
+                quant_config=quant_config,
+                prefix=f"{prefix}.down_proj",
+            )
+
+    def forward(self, x):
+        gate_up, _ = self.gate_up_proj(x)
+        x = self.act_fn(gate_up)
+        pad_size = 0
+        if self.enable_fc:
+            # pad input because AllGather requires token_num to be divisible by tp_size
+            x, pad_size = pad(x, self.tp_size)
+            output = torch.empty(x.shape, dtype=x.dtype, device=x.device)
+            dist.all_to_all_single(output, x)
+            x = output.reshape(self.tp_size, -1, output.size(-1)) \
+                        .transpose(0, 1) \
+                        .reshape(-1, output.size(-1)*self.tp_size)
+        x, _ = self.down_proj(x)
+        return x, pad_size
+
+
+class CustomQwen3Attention(Qwen3Attention):
+
+    def __init__(self,
+                 hidden_size: int,
+                 num_heads: int,
+                 num_kv_heads: int,
+                 max_position: int = 4096 * 32,
+                 head_dim: Optional[int] = None,
+                 rms_norm_eps: float = 1e-06,
+                 qkv_bias: bool = False,
+                 rope_theta: float = 10000,
+                 cache_config: Optional[CacheConfig] = None,
+                 quant_config: Optional[QuantizationConfig] = None,
+                 rope_scaling: Optional[tuple] = None,
+                 prefix: str = "",
+                 attn_type: str = AttentionType.DECODER) -> None:
+        super().__init__(hidden_size=hidden_size,
+                         num_heads=num_heads,
+                         num_kv_heads=num_kv_heads,
+                         max_position=max_position,
+                         head_dim=head_dim,
+                         rms_norm_eps=rms_norm_eps,
+                         qkv_bias=qkv_bias,
+                         rope_theta=rope_theta,
+                         cache_config=cache_config,
+                         quant_config=quant_config,
+                         rope_scaling=rope_scaling,
+                         prefix=prefix,
+                         attn_type=attn_type)
+        self.tp_size = get_tensor_model_parallel_world_size()
+        self.tp_rank = get_tensor_model_parallel_rank()
+        self.enable_fc = envs.VLLM_ENABLE_FC
+        if self.enable_fc:
+            self.o_proj = ReplicatedLinear(
+                self.total_num_heads * self.head_dim,
+                hidden_size,
+                bias=False,
+                quant_config=quant_config,
+                prefix=f"{prefix}.o_proj",
+            )
+        else:
+            self.o_proj = RowParallelLinear(
+                self.total_num_heads * self.head_dim,
+                hidden_size,
+                bias=False,
+                quant_config=quant_config,
+                prefix=f"{prefix}.o_proj",
+            )
+
+    def forward(
+        self,
+        positions: torch.Tensor,
+        hidden_states: torch.Tensor,
+    ) -> torch.Tensor:
+        qkv, _ = self.qkv_proj(hidden_states)
+        q, k, v = qkv.split([self.q_size, self.kv_size, self.kv_size], dim=-1)
+        # Add qk-norm
+        q_by_head = q.view(*q.shape[:-1], q.shape[-1] // self.head_dim,
+                           self.head_dim)
+        q_by_head = self.q_norm(q_by_head)
+        q = q_by_head.view(q.shape)
+        k_by_head = k.view(*k.shape[:-1], k.shape[-1] // self.head_dim,
+                           self.head_dim)
+        k_by_head = self.k_norm(k_by_head)
+        k = k_by_head.view(k.shape)
+        q, k = self.rotary_emb(positions, q, k)
+        attn_output = self.attn(q, k, v)
+        pad_size = 0
+        if self.enable_fc:
+            # pad input because AllGather requires token_num to be divisible by tp_size
+            attn_output, pad_size = pad(attn_output, self.tp_size)
+            output = torch.empty(attn_output.shape, dtype=attn_output.dtype, device=attn_output.device)
+            dist.all_to_all_single(output, attn_output)
+            attn_output = output.reshape(self.tp_size, -1, output.size(-1)) \
+                                .transpose(0, 1) \
+                                .reshape(-1, output.size(-1)*self.tp_size)
+        output, _ = self.o_proj(attn_output)
+        return output, pad_size
+
+
+class CustomQwen3DecoderLayer(nn.Module):
+
+    def __init__(
+        self,
+        config: Qwen3Config,
+        cache_config: Optional[CacheConfig] = None,
+        quant_config: Optional[QuantizationConfig] = None,
+        prefix: str = "",
+    ) -> None:
+        super().__init__()
+        self.hidden_size = config.hidden_size
+        self.tp_size = get_tensor_model_parallel_world_size()
+        self.tp_rank = get_tensor_model_parallel_rank()
+        self.enable_fc = envs.VLLM_ENABLE_FC
+        # Requires transformers > 4.32.0
+        rope_theta = getattr(config, "rope_theta", 1000000)
+        rope_scaling = getattr(config, "rope_scaling", None)
+
+        # By default, Qwen3 uses causal attention as it is a decoder-only model.
+        # You can override the HF config with `is_causal=False` to enable
+        # bidirectional attention, which is used in some embedding models
+        # (e.g. Alibaba-NLP/gte-Qwen3-7B-instruct)
+        if getattr(config, "is_causal", True):
+            attn_type = AttentionType.DECODER
+        else:
+            attn_type = AttentionType.ENCODER_ONLY
+
+        self.self_attn = CustomQwen3Attention(
+            hidden_size=self.hidden_size,
+            num_heads=config.num_attention_heads,
+            max_position=config.max_position_embeddings,
+            num_kv_heads=config.num_key_value_heads,
+            rope_theta=rope_theta,
+            rms_norm_eps=config.rms_norm_eps,
+            qkv_bias=getattr(config, 'attention_bias', False),
+            head_dim=getattr(config, 'head_dim', None),
+            cache_config=cache_config,
+            quant_config=quant_config,
+            rope_scaling=rope_scaling,
+            prefix=f"{prefix}.self_attn",
+            attn_type=attn_type,
+        )
+        self.mlp = CustomQwen3MLP(
+            hidden_size=self.hidden_size,
+            intermediate_size=config.intermediate_size,
+            hidden_act=config.hidden_act,
+            quant_config=quant_config,
+            prefix=f"{prefix}.mlp",
+        )
         if quant_config is None:
             return
 
@@ -56,6 +246,58 @@ def __init__(
                 layer=self.mlp.gate_up_proj,
                 eps=config.rms_norm_eps)
 
+    def pre_attention_process(self, hidden_states, residual, pad_size=0):
+        hidden_states, residual = self.input_layernorm(
+                hidden_states, residual)
+        hidden_states = tensor_model_parallel_all_gather(hidden_states, 0)
+        hidden_states = unpad(hidden_states, pad_size)
+        return hidden_states, residual
+    
+    def pre_mlp_process(self, hidden_states, residual, pad_size=0):
+        token_num = hidden_states.size(0)
+        if token_num != residual.size(0):
+            if pad_size > 0:
+                residual = F.pad(residual, (0, 0, 0, pad_size))
+            split_size_list = [token_num] * self.tp_size
+            residual = torch.split(residual, split_size_list)[self.tp_rank]
+
+        hidden_states, residual = self.post_attention_layernorm(
+            hidden_states, residual)
+        hidden_states = tensor_model_parallel_all_gather(hidden_states, 0)
+        hidden_states = unpad(hidden_states, pad_size)
+        return hidden_states, residual
+
+    def forward(
+        self,
+        positions: torch.Tensor,
+        hidden_states: torch.Tensor,
+        residual: Optional[torch.Tensor],
+        pad_size: int = 0
+    ) -> tuple[torch.Tensor, torch.Tensor]:
+        # Self Attention
+        if residual is None:
+            residual = hidden_states
+            hidden_states = self.input_layernorm(hidden_states)
+        else:
+            if self.enable_fc:
+                hidden_states, residual = self.pre_attention_process(hidden_states, residual, pad_size)
+            else:
+                hidden_states, residual = self.input_layernorm(
+                    hidden_states, residual)
+        hidden_states, pad_size = self.self_attn(
+            positions=positions,
+            hidden_states=hidden_states,
+        )
+
+        # Fully Connected
+        if self.enable_fc:
+            hidden_states, residual = self.pre_mlp_process(hidden_states, residual, pad_size)
+        else:
+            hidden_states, residual = self.post_attention_layernorm(
+                hidden_states, residual)
+        hidden_states, pad_size = self.mlp(hidden_states)
+        return hidden_states, residual, pad_size
+    
 
 ALL_DECODER_LAYER_TYPES = {
     "attention": CustomQwen3DecoderLayer,
@@ -77,6 +319,48 @@ def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         super().__init__(vllm_config=vllm_config,
                          prefix=prefix,
                          decoder_layer_type=CustomQwen3DecoderLayer)
+        self.tp_size = get_tensor_model_parallel_world_size()
+        self.tp_rank = get_tensor_model_parallel_rank()
+        self.enable_fc = envs.VLLM_ENABLE_FC
+
+    def forward(
+        self,
+        input_ids: torch.Tensor,
+        positions: torch.Tensor,
+        intermediate_tensors: Optional[IntermediateTensors] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+    ) -> Union[torch.Tensor, IntermediateTensors]:
+        if get_pp_group().is_first_rank:
+            if inputs_embeds is not None:
+                hidden_states = inputs_embeds
+            else:
+                hidden_states = self.get_input_embeddings(input_ids)
+            residual = None
+        else:
+            assert intermediate_tensors is not None
+            hidden_states = intermediate_tensors["hidden_states"]
+            residual = intermediate_tensors["residual"]
+        pad_size = 0
+        for layer in self.layers[self.start_layer:self.end_layer]:
+            hidden_states, residual, pad_size = layer(
+                positions,
+                hidden_states,
+                residual,
+                pad_size
+            )
+        if self.enable_fc:
+            hidden_states = tensor_model_parallel_all_gather(hidden_states, 0)
+            residual = tensor_model_parallel_all_gather(residual, 0)
+            if pad_size > 0:
+                hidden_states = hidden_states[:-pad_size]
+                residual = residual[:-pad_size]
+        if not get_pp_group().is_last_rank:
+            return IntermediateTensors({
+                "hidden_states": hidden_states,
+                "residual": residual
+            })
+        hidden_states, _ = self.norm(hidden_states, residual)
+        return hidden_states
 
 
 class CustomQwen3ForCausalLM(nn.Module, SupportsLoRA, SupportsPP):

Original file line number	Diff line number	Diff line change
`@@ -143,6 +143,8 @@`
`143`	`143`	`# Batch MC2 in prefill: The number of tokens in each batch`
`144`	`144`	`"VLLM_ASCEND_FUSED_MOE_MC2_CHUNK_SIZE":`
`145`	`145`	`lambda: int(os.getenv("VLLM_ASCEND_FUSED_MOE_MC2_CHUNK_SIZE", "128")),`
	`146`	`+ "VLLM_ENABLE_FC":`
	`147`	`+ lambda: int(os.getenv("VLLM_ENABLE_FC", 0))`
`146`	`148`	`}`
`147`	`149`
`148`	`150`	`# end-env-vars-definition`