Skip to content

[OpenCL] Fix a bug when running an external model with the image-loader transposing the input to NCHW #3951

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.

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

Closed
wants to merge 1 commit into from
Closed
Changes from all commits
Commits
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
8 changes: 8 additions & 0 deletions lib/Backends/OpenCL/OpenCLTensorLayout.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -44,6 +44,14 @@ static std::string returnBaseReqOrNHWC(TensorLayoutDescription &baseReq,
// These nodes accept any 4-D layout.
return baseReq.getSerializedLayout();
}
// For Placeholders and Constants that were loaded from another tool, we don't
// have the layout information in during load time. This makes us assume they
// are in Glow's Canonical NHWC format, which is not correct if we are loading
// an image with NCHW such as in resent loader. Weaken the verifier to avoid a
// runtime crash: if this is a storage location, return the baseReq.
if (llvm::dyn_cast<Storage>(node)) {
return baseReq.getSerializedLayout();
}

return CanonicalTensorLayout::getInstance().getDefaultNDLayout(4);
}
Expand Down