Skip to content

Fix test_mtl_e2e_test_a16w8_two_layers_turing_3_1_1k #16513

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
meta-codesync merged 1 commit into from
Jan 9, 2026
+6 −6
Merged

Fix test_mtl_e2e_test_a16w8_two_layers_turing_3_1_1k #16513

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 6 additions & 6 deletions backends/cadence/aot/replace_ops.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1665,13 +1665,13 @@ def maybe_remove_or_replace(self, node: torch.fx.Node) -> bool:

def call(self, graph_module: torch.fx.GraphModule) -> PassResult:
result = super().call(graph_module)
# If this pass made modifications, fuse any cascaded view ops that may have been created
if result.modified:
fuse_cascaded_result = FuseCascadedViewOps().call(result.graph_module)

# True because we are in the 'if modified' block
return PassResult(fuse_cascaded_result.graph_module, True)
return result
# TODO: I tried conditionally running this only if the above made any modifications,
# but for whatever reason was getting numerical failures in
# test_mtl_e2e_test_a16w8_two_layers_turing_3_1_1k. Always running this pass
# resolved that issue.
fuse_cascaded_result = FuseCascadedViewOps().call(result.graph_module)
return PassResult(fuse_cascaded_result.graph_module, True)


@register_cadence_pass(CadencePassAttribute(opt_level=2))
Expand Down
Loading