Skip to content

GitLab

Unverified Commit c3f05b09 authored by Chenguang Li's avatar Chenguang Li Committed by GitHub
Browse files

[Misc]Minor Changes about Worker (#11555) 


Signed-off-by: default avatarChenguang Li <757486878@qq.com>
parent cf6bbcb4
Hide whitespace changes
Inline Side-by-side
Showing with 0 additions and 2 deletions
vllm/v1/worker/gpu_worker.py
View file @ c3f05b09
...@@ -132,7 +132,6 @@ class Worker: ...@@ -132,7 +132,6 @@ class Worker:
# Execute a forward pass with dummy inputs to profile the memory usage # Execute a forward pass with dummy inputs to profile the memory usage
# of the model. # of the model.
self.model_runner.profile_run() self.model_runner.profile_run()
torch.cuda.synchronize()
free_gpu_memory, _ = torch.cuda.mem_get_info() free_gpu_memory, _ = torch.cuda.mem_get_info()
# NOTE(woosuk): Here we assume that the other processes using the same # NOTE(woosuk): Here we assume that the other processes using the same
......
vllm/worker/worker.py
View file @ c3f05b09
...@@ -200,7 +200,6 @@ class Worker(LocalOrDistributedWorkerBase): ...@@ -200,7 +200,6 @@ class Worker(LocalOrDistributedWorkerBase):
weights_memory_in_bytes=self.model_runner. weights_memory_in_bytes=self.model_runner.
model_memory_usage) as result: model_memory_usage) as result:
self.model_runner.profile_run() self.model_runner.profile_run()
torch.cuda.synchronize()
self._assert_memory_footprint_increased_during_profiling() self._assert_memory_footprint_increased_during_profiling()
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment