[Bugfix] Fix size calculation of processing cache

tests/multimodal/test_processing.py

@@ -7,15 +7,20 @@
 
 import numpy as np
 import pytest
+import torch
 from transformers import ProcessorMixin
 
 from vllm.config import ModelConfig
 from vllm.multimodal import MULTIMODAL_REGISTRY
+from vllm.multimodal.inputs import (MultiModalFieldElem, MultiModalKwargs,
+                                    MultiModalKwargsItem,
+                                    MultiModalSharedField)
 # yapf conflicts with isort for this block
 # yapf: disable
 from vllm.multimodal.processing import (PlaceholderFeaturesInfo,
-                                        PromptIndexTargets, PromptInsertion,
-                                        PromptReplacement, apply_text_matches,
+                                        ProcessingCache, PromptIndexTargets,
+                                        PromptInsertion, PromptReplacement,
+                                        apply_text_matches,
                                         apply_token_matches,
                                         find_mm_placeholders,
                                         find_text_matches, find_token_matches,
@@ -890,6 +895,45 @@ def test_find_mm_placeholders(
     assert result == expected
 
 
+def _dummy_elem(modality: str, key: str, size: int):
+    return MultiModalFieldElem(
+        modality=modality,
+        key=key,
+        data=torch.empty((size, ), dtype=torch.int8),
+        field=MultiModalSharedField(1),
+    )
+
+
+def _dummy_item(modality: str, size_by_key: dict[str, int]):
+    return MultiModalKwargsItem.from_elems([
+        _dummy_elem(modality, key, size) for key, size in size_by_key.items()
+    ])
+
+
+def _dummy_kw(size_by_key_modality: dict[str, dict[str, int]]):
+    return MultiModalKwargs.from_items([
+        _dummy_item(modality, size_by_key)
+        for modality, size_by_key in size_by_key_modality.items()
+    ])
+
+
+# yapf: disable
+@pytest.mark.parametrize(
+    ("item", "expected_size"),
+    [
+        (_dummy_item("a", {"a1": 100}), 100),
+        (_dummy_item("a", {"a1": 100, "a2": 110}), 210),
+        (_dummy_kw({"a": {"a1": 100, "a2": 110}, "b": {"b1": 120, "b2": 130}}), 460),  # noqa: E501
+    ],
+)
+# yapf: enable
+def test_cache_item_size(item, expected_size):
+    cache = ProcessingCache.get_lru_cache(2048, type(item))
+    cache[""] = item
+
+    assert cache.currsize == expected_size
+
+
 @pytest.mark.parametrize("model_id", ["llava-hf/llava-v1.6-mistral-7b-hf"])
 @pytest.mark.parametrize(
     ("limit", "num_supported", "is_valid"),

vllm/multimodal/processing.py

@@ -26,7 +26,7 @@
 from .hasher import MultiModalHasher
 from .inputs import (MultiModalDataDict, MultiModalEncDecInputs,
                      MultiModalFieldConfig, MultiModalInputs, MultiModalKwargs,
-                     MultiModalKwargsItem, PlaceholderRange)
+                     MultiModalKwargsItem, NestedTensors, PlaceholderRange)
 from .parse import (DictEmbeddingItems, EmbeddingItems, MultiModalDataItems,
                     MultiModalDataParser)
 
@@ -853,33 +853,62 @@ class ProcessingCache:
 
     @staticmethod
     def get_lru_cache(
-        capacity_gb: int,
+        capacity_gb: float,
         value_type: type[_V],
+        *,
+        debug: bool = False,
     ) -> LRUCache[str, _V]:
 
-        def get_size(leaf: object) -> int:
+        def get_leaf_size(leaf: object) -> int:
+            # MultiModalKwargs is not a subclass of dict
+            if isinstance(leaf, MultiModalKwargs):
+                return get_item_size(leaf.data)
+
+            # MultiModalKwargsItem is not a subclass of dict
+            if isinstance(leaf, MultiModalKwargsItem):
+                leaf_data = {k: v.data for k, v in leaf.items()}
+                return get_item_size(leaf_data)
+
+            # sys.getsizeof doesn't work for tensors
             if isinstance(leaf, torch.Tensor):
-                return leaf.nbytes  # sys.getsizeof doesn't work for tensors
+                return leaf.nbytes
 
             return sys.getsizeof(leaf)
 
-        return LRUCache[str, _V](
-            GiB_bytes * capacity_gb,
-            getsizeof=lambda x: json_reduce_leaves(
+        def get_item_size(
+            value: Union[MultiModalKwargs, MultiModalKwargsItem,
+                         Mapping[str, NestedTensors]]
+        ) -> int:
+            size = json_reduce_leaves(
                 lambda a, b: a + b,
-                json_map_leaves(get_size, x),
-            ),
-        )
+                json_map_leaves(get_leaf_size, value),
+            )
+
+            if debug:
+                logger.debug("Calculated size of %s to be %.2f GiB",
+                             type(value), size / GiB_bytes)
 
-    def __init__(self, capacity_gb: int) -> None:
+            return size
+
+        return LRUCache(GiB_bytes * capacity_gb, getsizeof=get_item_size)
+
+    def __init__(
+        self,
+        capacity_gb: float,
+        *,
+        debug_cache_hit_ratio_steps: Optional[int] = None,
+    ) -> None:
         super().__init__()
 
-        # DEBUG: Set to None to disable
-        self.debug_cache_hit_ratio_steps: Optional[int] = None
+        self.debug_cache_hit_ratio_steps = debug_cache_hit_ratio_steps
         self.debug_cache_hits = 0
         self.debug_cache_total = 0
 
-        self._cache = self.get_lru_cache(capacity_gb, MultiModalKwargsItem)
+        self._cache = self.get_lru_cache(
+            capacity_gb,
+            MultiModalKwargsItem,
+            debug=bool(debug_cache_hit_ratio_steps),
+        )
 
     def _maybe_log_cache_stats(self) -> None:
         steps = self.debug_cache_hit_ratio_steps
@@ -890,6 +919,9 @@ def _maybe_log_cache_stats(self) -> None:
         if total > 0 and total % steps == 0:
             logger.debug("ProcessingCache: hit_ratio = %.2f",
                          self.debug_cache_hits / total)
+            logger.debug("ProcessingCache: size = %.2f / %.2f GiB",
+                         self._cache.currsize / GiB_bytes,
+                         self._cache.maxsize / GiB_bytes)
 
     def get(
         self,