Check the validity of the buffer before parsing the model

If the buffer does not point to a well defined flatbuffer the parsing
segfaults.

Change-Id: Icb8dfef37dc28b2b7a22c6d3804851be8198aa9d

3 files changed, 24 insertions, 8 deletions

diff --git a/applications/inference_process/include/inference_parser.hpp b/applications/inference_process/include/inference_parser.hpp
index 3d90818..fd7a97a 100644
--- a/applications/inference_process/include/inference_parser.hpp
+++ b/applications/inference_process/include/inference_parser.hpp
@@ -56,15 +56,30 @@ Array<T, U> makeArray(T *const data, U &size, size_t capacity) {
 
 class InferenceParser {
 public:
-    template <typename T, typename U, size_t S>
-    bool parseModel(const void *buffer, char (&description)[S], T &&ifmDims, U &&ofmDims) {
+    const tflite::Model *getModel(const void *buffer, size_t size) {
+        // Verify buffer
+        flatbuffers::Verifier base_verifier(reinterpret_cast<const uint8_t *>(buffer), size);
+        if (!tflite::VerifyModelBuffer(base_verifier)) {
+            printf("Warning: the model is not valid\n");
+            return nullptr;
+        }
+
         // Create model handle
         const tflite::Model *model = tflite::GetModel(buffer);
         if (model->subgraphs() == nullptr) {
             printf("Warning: nullptr subgraph\n");
-            return true;
+            return nullptr;
         }
 
+        return model;
+    }
+
+    template <typename T, typename U, size_t S>
+    bool parseModel(const void *buffer, size_t size, char (&description)[S], T &&ifmDims, U &&ofmDims) {
+        const tflite::Model *model = getModel(buffer, size);
+        if (model == nullptr) {
+            return true;
+        }
         strncpy(description, model->description()->c_str(), sizeof(description));
 
         // Get input dimensions for first subgraph
diff --git a/applications/inference_process/include/inference_process.hpp b/applications/inference_process/include/inference_process.hpp
index fc54ae0..f8d7fd8 100644
--- a/applications/inference_process/include/inference_process.hpp
+++ b/applications/inference_process/include/inference_process.hpp
@@ -18,6 +18,8 @@
 
 #pragma once
 
+#include "inference_parser.hpp"
+
 #include <array>
 #include <queue>
 #include <stdlib.h>
@@ -85,5 +87,6 @@ private:
 
     uint8_t *tensorArena;
     const size_t tensorArenaSize;
+    InferenceParser parser;
 };
 } // namespace InferenceProcess
diff --git a/applications/inference_process/src/inference_process.cpp b/applications/inference_process/src/inference_process.cpp
index 29254c7..e96d601 100644
--- a/applications/inference_process/src/inference_process.cpp
+++ b/applications/inference_process/src/inference_process.cpp
@@ -119,11 +119,9 @@ bool InferenceProcess::runJob(InferenceJob &job) {
     RegisterDebugLogCallback(tfluDebugLog);
 
     // Get model handle and verify that the version is correct
-    const tflite::Model *model = ::tflite::GetModel(job.networkModel.data);
-    if (model->version() != TFLITE_SCHEMA_VERSION) {
-        LOG_ERR("Model schema version unsupported: version=%" PRIu32 ", supported=%d.",
-                model->version(),
-                TFLITE_SCHEMA_VERSION);
+    const tflite::Model *model = parser.getModel(job.networkModel.data, job.networkModel.size);
+    if (model == nullptr) {
+        LOG_ERR("Invalid model");
         return true;
     }