[Misc] Add numpy implementation of compute_slot_mapping (#7377)

vllm/attention/backends/utils.py

@@ -1,6 +1,7 @@
 """Attention backend utils"""
 from typing import TYPE_CHECKING, Dict, List, Type, TypeVar, Union
 
+import numpy as np
 import torch
 
 from vllm.attention import AttentionMetadata, AttentionMetadataBuilder
@@ -13,6 +14,10 @@ STR_NOT_IMPL_ENC_DEC_ROCM_HIP = ("ROCm/HIP is not currently supported "
 
 PAD_SLOT_ID = -1
 
+# Switch to numpy implementation of compute_slot_mapping
+# if we have at least this many elements. Could be tuned further.
+_COMPUTE_SLOT_MAPPING_NUMPY_NUMEL = 256
+
 if TYPE_CHECKING:
     from vllm.worker.model_runner import ModelInputForGPUBuilder
 
@@ -46,6 +51,29 @@ def compute_slot_mapping_start_idx(is_prompt: bool, query_len: int,
     return start_idx
 
 
+def _compute_slot_mapping_python(slot_mapping: List[int],
+                                 block_table: List[int], range_start: int,
+                                 range_end: int, block_size: int):
+    for i in range(range_start, range_end):
+        block_number = block_table[i // block_size]
+        block_offset = i % block_size
+        slot = block_number * block_size + block_offset
+        slot_mapping.append(slot)
+
+
+def _compute_slot_mapping_numpy(slot_mapping: List[int],
+                                block_table: List[int], range_start: int,
+                                range_end: int, block_size: int):
+    block_table_array = np.array(block_table)
+    idx = np.arange(range_start, range_end)
+    block_offset = idx % block_size
+    idx //= block_size
+    seq_slot_mapping_array = block_table_array[idx]
+    seq_slot_mapping_array *= block_size
+    seq_slot_mapping_array += block_offset
+    slot_mapping.extend(seq_slot_mapping_array)
+
+
 def compute_slot_mapping(is_profile_run: bool, slot_mapping: List[int],
                          seq_id: int, seq_len: int, context_len: int,
                          start_idx: int, block_size: int,
@@ -67,21 +95,22 @@ def compute_slot_mapping(is_profile_run: bool, slot_mapping: List[int],
     # sliding window is 8, and block size is 4, the first two
     # tokens are masked and the slot mapping will be
     # [-1, -1, 2, 3, 4, 5, 6, 7, 0, 1].
+    padding_mask_len = max(0, start_idx - context_len)
+    slot_mapping.extend([PAD_SLOT_ID] * padding_mask_len)
+
+    range_start = max(start_idx, context_len)
+    range_end = seq_len
+    numel = range_end - range_start
     block_table = block_tables[seq_id]
 
-    def add_slot(i):
-        block_number = block_table[i // block_size]
-        block_offset = i % block_size
-        slot = block_number * block_size + block_offset
-        slot_mapping.append(slot)
-
-    if start_idx == 0 and (seq_len - context_len) == 1:
-        # Optimization for common-case of decoding next token
-        add_slot(seq_len - 1)
+    # numpy implementation will be faster than python if we have
+    # many elements, otherwise it will be slower.
+    if numel < _COMPUTE_SLOT_MAPPING_NUMPY_NUMEL:
+        _compute_slot_mapping_python(slot_mapping, block_table, range_start,
+                                     range_end, block_size)
     else:
-        slot_mapping.extend([PAD_SLOT_ID] * max(0, start_idx - context_len))
-        for i in range(max(start_idx, context_len), seq_len):
-            add_slot(i)
+        _compute_slot_mapping_numpy(slot_mapping, block_table, range_start,
+                                    range_end, block_size)
 
 
 TAttentionMetadata = TypeVar("TAttentionMetadata", bound='AttentionMetadata')