Skip unlink logic if load/offload dev match
i.e. when forcing it to run on CPU or highvram enabled
This commit is contained in:
@@ -99,7 +99,7 @@ class GGUFModelPatcher(comfy.model_patcher.ModelPatcher):
|
||||
if device == self.offload_device:
|
||||
linked.append((n, m))
|
||||
continue
|
||||
if linked:
|
||||
if linked and self.load_device != self.offload_device:
|
||||
logging.info(f"Attempting to release mmap ({len(linked)})")
|
||||
for n, m in linked:
|
||||
# TODO: possible to OOM, find better way to detach
|
||||
|
||||
Reference in New Issue
Block a user