Skip to content

fix: Treat empty DLPack tensors as contiguous #444

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

Open
nsiddharth wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: Treat empty DLPack tensors as contiguous #444

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
16 changes: 14 additions & 2 deletions src/pb_tensor.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -493,8 +493,20 @@ PbTensor::FromDLPackCapsule(
dl_managed_tensor->dl_tensor.shape,
dl_managed_tensor->dl_tensor.shape + ndim);

// Check if the input is contiguous and in C order
if (strides != nullptr) {
// Check if the input is contiguous and in C order.
// An empty tensor (any zero-size dimension, i.e. zero elements) has no
// meaningful memory layout and is trivially contiguous; skip the stride
// check, which can otherwise reject it because frameworks report
// inconsistent strides for empty tensors. See:
// https://github.com/triton-inference-server/server/issues/6960
bool is_empty = false;
for (auto dim : dims) {
if (dim == 0) {
is_empty = true;
break;
}
}
if (strides != nullptr && !is_empty) {
int64_t calculated_stride{1};
bool is_contiguous_c_order = true;
for (size_t i = 1; i < dims.size(); i++) {
Expand Down