biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[torch.compile] Cleanup compilation tests and custom passes, add debug utils, fix DCE bug (#23091), fix test (#24376), and prep for custom op matching (#24604) (#24542)

Browse Source 
Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Signed-off-by: luka <lgovedic@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
This commit is contained in:
Luka Govedič
2025-09-22 15:30:05 -04:00
committed by GitHub
parent 8d0ee5a564
commit d5e0fca264
24 changed files with 404 additions and 461 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/compile/test_async_tp.py
View File

@@ -294,6 +294,8 @@ def async_tp_pass_on_test_model(local_rank: int, world_size: int,
compiled_model = torch.compile(model, backend=backend)
compiled_model(hidden_states)
assert async_tp_pass.matched_count == 1
# In pre-nodes, all gather or reduce scatter should exist,
# fused_matmul_reduce_scatter or fused_all_gather_matmul should not
backend.check_before_ops(model.ops_in_model_before(), fully_replaced=False)