Update file reading function to fix external data loading (#1460)

src/file_buffer.cpp

@@ -30,23 +30,31 @@ namespace migraphx {
 inline namespace MIGRAPHX_INLINE_NS {
 
 template <class T>
-T generic_read_file(const std::string& filename)
+T generic_read_file(const std::string& filename, size_t offset = 0, size_t nbytes = 0)
 {
     std::ifstream is(filename, std::ios::binary | std::ios::ate);
-    std::streamsize size = is.tellg();
-    if(size < 1)
+    if(nbytes == 0)
+    {
+        // if there is a non-zero offset and nbytes is not set,
+        // calculate size of remaining bytes to read
+        nbytes = is.tellg();
+        if(offset > nbytes)
+            MIGRAPHX_THROW("offset is larger than file size");
+        nbytes -= offset;
+    }
+    if(nbytes < 1)
         MIGRAPHX_THROW("Invalid size for: " + filename);
-    is.seekg(0, std::ios::beg);
+    is.seekg(offset, std::ios::beg);
 
-    T buffer(size, 0);
-    if(not is.read(&buffer[0], size))
+    T buffer(nbytes, 0);
+    if(not is.read(&buffer[0], nbytes))
         MIGRAPHX_THROW("Error reading file: " + filename);
     return buffer;
 }
 
-std::vector<char> read_buffer(const std::string& filename)
+std::vector<char> read_buffer(const std::string& filename, size_t offset, size_t nbytes)
 {
-    return generic_read_file<std::vector<char>>(filename);
+    return generic_read_file<std::vector<char>>(filename, offset, nbytes);
 }
 
 std::string read_string(const std::string& filename)

src/include/migraphx/file_buffer.hpp

@@ -31,7 +31,7 @@
 namespace migraphx {
 inline namespace MIGRAPHX_INLINE_NS {
 
-std::vector<char> read_buffer(const std::string& filename);
+std::vector<char> read_buffer(const std::string& filename, size_t offset = 0, size_t nbytes = 0);
 std::string read_string(const std::string& filename);
 
 void write_buffer(const std::string& filename, const char* buffer, std::size_t size);

src/onnx/onnx_parser.cpp

@@ -393,18 +393,31 @@ literal onnx_parser::parse_value(const onnx::AttributeProto& attr) const
 literal onnx_parser::parse_tensor(const onnx::TensorProto& t) const
 {
     std::vector<std::size_t> dims(t.dims().begin(), t.dims().end());
-    if(not t.external_data().empty())
+    auto type = get_type(t.data_type());
+    shape tensor_shape(type, dims);
+    auto external_data = t.external_data();
+    if(not external_data.empty())
     {
-        const std::string& data_file = t.external_data().at(0).value();
-        auto raw_buffer              = read_buffer(path + "/" + data_file);
+        const std::string& data_file = external_data.at(0).value();
+        size_t num_data_fields       = external_data.size();
+        size_t offset                = 0;
+        size_t nbytes                = tensor_shape.bytes();
+
+        if(num_data_fields > 1) // if offset field is present
+        {
+            offset = std::stoul(t.external_data().at(1).value());
+        }
+        if(num_data_fields > 2) // if nbytes field is present
+        {
+            nbytes = std::stoul(t.external_data().at(2).value());
+        }
+        auto raw_buffer = read_buffer(path + "/" + data_file, offset, nbytes);
         std::string s(raw_buffer.begin(), raw_buffer.end());
-        auto type = get_type(t.data_type());
         return create_literal(type, dims, s.data());
     }
     if(t.has_raw_data())
     {
         const std::string& s = t.raw_data();
-        auto type            = get_type(t.data_type());
         return create_literal(type, dims, s.data());
     }
 

test/onnx/external_constant_test.onnx

+external_constant_test:¡

+v
0"Constant*g
+value*[Bconst_tensorj)
+locationexternal_constant_test.weightj
+offset48j
+length24p
 
external_constant_testb
+
0
+
+

+B
\ No newline at end of file

test/onnx/onnx_test.cpp

@@ -1818,6 +1818,16 @@ migraphx::program create_external_data_prog()
     return p;
 }
 
+TEST_CASE(external_constant_test)
+{
+    migraphx::program p;
+    auto* mm = p.get_main_module();
+    mm->add_literal(migraphx::literal{{migraphx::shape::int64_type, {3}}, {0, 1, 2}});
+
+    auto prog = optimize_onnx("external_constant_test.onnx");
+    EXPECT(p == prog);
+}
+
 TEST_CASE(external_data_test)
 {
     migraphx::program p = create_external_data_prog();