Skip to content

LLM inference in C/C++

C++ ggml

ggml-zdnn: fix 0-row tensor crash (#29636)

Aaron Teo86ea01d05ec2
diff --git a/ggml/src/ggml-zdnn/ggml-zdnn.cpp b/ggml/src/ggml-zdnn/ggml-zdnn.cpp
index 4007ac9dfc..3b45370a8a 100644
--- a/ggml/src/ggml-zdnn/ggml-zdnn.cpp
+++ b/ggml/src/ggml-zdnn/ggml-zdnn.cpp
@@ -225,6 +225,11 @@ static enum ggml_status ggml_backend_zdnn_buffer_init_tensor(ggml_backend_buffer
return GGML_STATUS_SUCCESS;
}
+ // reject empty tensors to avoid zDNN crash
+ if (ggml_is_empty(tensor)) {
+ return GGML_STATUS_SUCCESS;
+ }
+
ggml_backend_zdnn_buffer_context * ctx = (ggml_backend_zdnn_buffer_context *)buffer->context;
const int64_t tsize = ggml_nbytes(tensor);