Skip to content

Fix subscript simplification in C codegen #1036

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
inducer merged 1 commit into from
Aug 12, 2026
Merged

Fix subscript simplification in C codegen #1036

Show file tree
Hide file tree
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
2 changes: 1 addition & 1 deletion doc/tutorial.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1919,7 +1919,7 @@ Now to make things more interesting, we'll create a kernel with barriers:
barrier(CLK_LOCAL_MEM_FENCE) /* for c (insn rev-depends on insn_0) */;
c[990 * i + 99 * j + lid(0) + 1] = 2 * a[980 * i + 98 * j + lid(0) + 1];
barrier(CLK_LOCAL_MEM_FENCE) /* for c (insn_0 depends on insn) */;
e[980 * i + 98 * j + lid(0) + 1] = c[990 * i + 99 * j + 1 + lid(0) + 1] + c[990 * i + 99 * j + -1 + lid(0) + 1];
e[980 * i + 98 * j + lid(0) + 1] = c[1 + 990 * i + 99 * j + lid(0) + 1] + c[-1 + 990 * i + 99 * j + lid(0) + 1];
}
}

Expand Down
5 changes: 3 additions & 2 deletions loopy/target/c/codegen/expression.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,7 @@
LocalHardwareAxisIndex,
ResolvedFunction,
SubArrayRef,
flatten,
)
from loopy.target.c import CExpression
from loopy.type_inference import TypeInferenceMapper, TypeReader
Expand Down Expand Up @@ -336,8 +337,8 @@ def make_var(name):
result = self.make_subscript(
ary,
make_var(access_info.array_name),
simplify_using_aff(
self.kernel, self.rec_arith(subscript, "i")))
flatten(self.rec_arith(
simplify_using_aff(self.kernel, subscript), "i")))

if access_info.vector_index is not None:
return self.codegen_state.ast_builder.add_vector_access(
Expand Down
Loading