[Hardware] Replace torch.cuda.synchronize() api with torch.accelerator.synchronize (#36085)

Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
This commit is contained in:
Kunshang Ji
2026-03-05 10:36:39 +00:00
committed by GitHub
parent 0bfa229bf1
commit 66a2209645
59 changed files with 158 additions and 161 deletions
+1 -1
View File
@@ -622,7 +622,7 @@ def rearrange_expert_weights_inplace(
# NOTE(bowen): We need this synchronize to run, but I don't know why.
# If you figure out the reason, please let me know -- thank you!
torch.cuda.synchronize()
torch.accelerator.synchronize()
old_global_expert_indices_cpu = old_global_expert_indices.cpu().numpy()
new_global_expert_indices_cpu = new_global_expert_indices.cpu().numpy()