[Misc][Bugfix] FA3 support to ViT MHA layer (vllm-project#12435)

Signed-off-by: Roger Wang <[email protected]>
Signed-off-by: Isotr0py <[email protected]>
Co-authored-by: Isotr0py <[email protected]>

Author: 2 people

vllm/attention/layer.py

@@ -251,9 +251,28 @@ def forward(
                 _Backend.FLASH_ATTN,
                 _Backend.FLASH_ATTN_VLLM_V1,
         }:
-            from vllm.vllm_flash_attn import flash_attn_func
-
-            out = flash_attn_func(query, key, value, softmax_scale=self.scale)
+            from vllm.vllm_flash_attn import flash_attn_varlen_func
+
+            cu_seqlens_q = torch.arange(0, (bsz + 1) * q_len,
+                                        step=q_len,
+                                        dtype=torch.int32,
+                                        device=query.device)
+            cu_seqlens_k = torch.arange(0, (bsz + 1) * kv_len,
+                                        step=kv_len,
+                                        dtype=torch.int32,
+                                        device=key.device)
+
+            out = flash_attn_varlen_func(
+                query.flatten(0, 1),
+                key.flatten(0, 1),
+                value.flatten(0, 1),
+                cu_seqlens_q=cu_seqlens_q,
+                cu_seqlens_k=cu_seqlens_k,
+                max_seqlen_q=q_len,
+                max_seqlen_k=kv_len,
+                softmax_scale=self.scale,
+            )
+            out = out.reshape(bsz, q_len, -1)
         elif self.attn_backend == _Backend.XFORMERS:
             from xformers import ops as xops