Skip to content

Fix bug found by asan from #5784 #5798

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
dsharletg merged 1 commit into from
Mar 10, 2021
Merged

Fix bug found by asan from #5784 #5798

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
5 changes: 3 additions & 2 deletions src/CodeGen_LLVM.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1950,9 +1950,10 @@ void CodeGen_LLVM::visit(const Load *op) {
// LLVM misses optimizations like using ldN on ARM.
vector<Value *> results;
for (int i = 0; i < op->type.lanes(); i += slice_lanes) {
int load_lanes_i = std::min<int>(slice_lanes * stride->value, load_lanes - i);
int load_base_i = i * stride->value;
int load_lanes_i = std::min<int>(slice_lanes * stride->value, load_lanes - load_base_i);
int lanes_i = std::min<int>(slice_lanes, op->type.lanes() - i);
Expr slice_base = simplify(base + i * ramp->stride);
Expr slice_base = simplify(base + load_base_i);

Value *load_i = codegen_dense_vector_load(op->type.with_lanes(load_lanes_i), op->name, slice_base,
op->image, op->param, op->alignment, nullptr, false);
Expand Down