Add use of fused_partial_rope op (#10942)

Author: lshpku

paddlenlp/transformers/deepseek_v2/modeling.py

@@ -106,6 +106,11 @@ def swiglu(x, y=None):
             x, y = paddle.chunk(x, chunks=2, axis=-1)
         return F.silu(x) * y
 
+try:
+    from paddle.incubate.nn.functional import fused_partial_rope
+except ImportError:
+    fused_partial_rope = None
+
 
 __all__ = [
     "DeepseekV2LMHead",
@@ -1089,7 +1094,7 @@ def repeat_kv(hidden_states: paddle.Tensor, n_rep: int) -> paddle.Tensor:
 
 
 @to_static(backend="CINN")
-def qkv_pre_process(
+def qkv_pre_process_no_fuse(
     q, kv, k_pe, rotary_emb, num_heads, q_head_dim, qk_nope_head_dim, v_head_dim, qk_rope_head_dim, position_ids
 ):
     bsz, q_len, _ = q.shape
@@ -1125,6 +1130,50 @@ def qkv_pre_process(
     return query_states, key_states, value_states
 
 
+@to_static(backend="CINN")
+def rearrange_kv(kv, k_pe, qk_nope_head_dim, num_heads):
+    k_nope = kv[..., :qk_nope_head_dim]
+    value_states = kv[..., qk_nope_head_dim:]
+
+    k_pe = k_pe.expand([k_pe.shape[0], k_pe.shape[1], num_heads, k_pe.shape[3]])
+    key_states = paddle.concat([k_nope, k_pe], axis=-1)
+
+    return key_states, value_states
+
+
+def qkv_pre_process(
+    q, kv, k_pe, rotary_emb, num_heads, q_head_dim, qk_nope_head_dim, v_head_dim, qk_rope_head_dim, position_ids
+):
+    if (fused_partial_rope is None) or (position_ids is not None):
+        return qkv_pre_process_no_fuse(
+            q, kv, k_pe, rotary_emb, num_heads, q_head_dim, qk_nope_head_dim, v_head_dim, qk_rope_head_dim, position_ids
+        )
+
+    bsz, q_len, _ = q.shape
+
+    target_query_shape = [0, 0, num_heads, q_head_dim]
+    target_key_value_shape = [0, 0, num_heads, qk_nope_head_dim + v_head_dim]
+
+    q = q.reshape(shape=target_query_shape)
+    kv = kv.reshape(shape=target_key_value_shape)
+    k_pe = k_pe.reshape([-1, q_len, 1, qk_rope_head_dim])
+
+    value_states = kv[..., qk_nope_head_dim:]
+
+    kv_seq_len = value_states.shape[1]
+
+    cos, sin = rotary_emb(value_states, seq_len=kv_seq_len)
+    cos = cos[None, :, None, :]
+    sin = sin[None, :, None, :]
+
+    query_states = fused_partial_rope(q, cos, sin)
+    k_pe = fused_partial_rope(k_pe, cos, sin)
+
+    key_states, value_states = rearrange_kv(kv, k_pe, qk_nope_head_dim, num_heads)
+
+    return query_states, key_states, value_states
+
+
 def manul_fwd(
     q_init,
     kv_init,