[sleep mode] save memory for on-the-fly quantization (#24731)

Signed-off-by: youkaichao <youkaichao@gmail.com>

vllm/device_allocator/cumem.py

@@ -16,8 +16,11 @@ from typing import Any, Callable, Optional, Union
 
 import torch
 
+from vllm.logger import init_logger
 from vllm.utils import is_pin_memory_available
 
+logger = init_logger(__name__)
+
 
 def find_loaded_library(lib_name) -> Optional[str]:
     """
@@ -165,6 +168,9 @@ class CuMemAllocator:
         py_d_mem = allocation_handle[2]
         self.pointer_to_data[py_d_mem] = AllocationData(
             allocation_handle, self.current_tag)
+        logger.debug(
+            "Allocated %s bytes for %s with address %s from cumem allocator",
+            allocation_handle[1], self.current_tag, py_d_mem)
         return
 
     def _python_free_callback(self, ptr: int) -> HandleType:
@@ -174,6 +180,9 @@ class CuMemAllocator:
         data = self.pointer_to_data.pop(ptr)
         if data.cpu_backup_tensor is not None:
             data.cpu_backup_tensor = None
+        logger.debug(
+            "Freed %s bytes for %s with address %s from cumem allocator",
+            data.handle[1], data.tag, ptr)
         return data.handle
 
     def sleep(
@@ -197,9 +206,14 @@ class CuMemAllocator:
 
         assert isinstance(offload_tags, tuple)
 
+        total_bytes = 0
+        backup_bytes = 0
+
         for ptr, data in self.pointer_to_data.items():
             handle = data.handle
+            total_bytes += handle[1]
             if data.tag in offload_tags:
+                backup_bytes += handle[1]
                 size_in_bytes = handle[1]
                 cpu_backup_tensor = torch.empty(
                     size_in_bytes,
@@ -211,6 +225,12 @@ class CuMemAllocator:
                 data.cpu_backup_tensor = cpu_backup_tensor
             unmap_and_release(handle)
 
+        logger.info(
+            "CuMemAllocator: sleep freed %.2f GiB memory in total, of which "
+            "%.2f GiB is backed up in CPU and the rest %.2f GiB is discarded "
+            "directly.", total_bytes / 1024**3, backup_bytes / 1024**3,
+            (total_bytes - backup_bytes) / 1024**3)
+
         gc.collect()
         torch.cuda.empty_cache()
 
@@ -267,12 +287,17 @@ class CuMemAllocator:
             # when using pluggable allocator, see
             # https://github.com/pytorch/pytorch/issues/145168 .
             # if we have some memory allocated and then freed,
-            # the memory will not be released.
+            # the memory will not be released, e.g. in online quantization,
-            # right now it is fine, because we only use this allocator
+            # where the model is created in higher precision, and then
-            # during weight loading and kv cache creation, where we only
+            # quantized in lower precision.
-            # allocate memory.
+            # Find all unused allocations and manually release them.
-            # TODO: we need to find a way to release the memory,
+            # TODO: we should expose `empty_cache` method in the memory pool.
-            # i.e. calling torch.cuda.empty_cache()
+            # TODO: ask for help from PyTorch team to expose this method.
+            allocations = data[0].snapshot()
+            for allocation in allocations:
+                if allocation["allocated_size"] == 0:
+                    handle = self._python_free_callback(allocation["address"])
+                    unmap_and_release(handle)
             self.current_tag = old_tag
 
     def get_current_usage(self) -> int: