I have an one question when I use C# DllImport C++ dll, I use the visual studio 2010 & checked the "Enable unmanaged code debugging", when it's running, always show the message "Buffer overrun detected! ... A buffer overrun has been detected which has corrupted the program's internal state. The program cannot safely continue execution and must now be terminated."
This dll is Third-party vendors to provide, and they says it's no error to run this dll, how can i fixed it?
My M++ dll function is,
int avc_to_avi_convert(char* chPath_avc, char* chPath_avi, int pPrivate, PROGRESS_CALLBACK progress_callback);
void avc_to_avi_close(int* pFd_avi);
typedef void (*PROGRESS_CALLBACK)(int iPercent, int pPrivate);
And i am use it in my C# dllimport like this :
public delegate void PROGRESS_CALLBACK(int _iPercent, int _pPrivate);
[DllImportAttribute(#"..\xxx.dll", EntryPoint = "avc_to_avi_convert", CallingConvention = CallingConvention.Cdecl, CharSet = CharSet.Unicode)]
public static extern unsafe int avc_to_avi_convert([MarshalAs(UnmanagedType.LPStr)] StringBuilder _inputAVC, [MarshalAs(UnmanagedType.LPStr)] StringBuilder _ouputAVI, int pPrivate, PROGRESS_CALLBACK _pg);
[DllImportAttribute(#"..\xxx.dll", EntryPoint = "avc_to_avi_close", CallingConvention = CallingConvention.Cdecl, CharSet = CharSet.Unicode)]
public static extern unsafe void avc_to_avi_close(int pFd_avi);
private void button1_Click(object sender, EventArgs e)
{
int i = 0;
StringBuilder inputFile = new StringBuilder(Application.StartupPath + #"\avc\abc.avc");//(#"C:\avc\abc.avc");
StringBuilder outputFile = new StringBuilder(Application.StartupPath + #"\avi\abc.avi");//(#"C:\avi\abc.avi");
if (avc_to_avi_convert(
inputFile,
outputFile,
1,
progress_func) > 0) {
}
}
public void progress_func(int iProgress, int pPrivate)
{
if (iProgress == 100)
{
//success
}
else if (iProgress == -1)
{
//convert error
}
else if (iProgress == -2)
{
//Not enough disk space
}
else
{
//update progress
}
}
I changed my code to,
[DllImportAttribute(#"..\xxx.dll", EntryPoint = "avc_to_avi_convert", CallingConvention = CallingConvention.StdCall, CharSet = CharSet.Unicode)]
public static extern unsafe int avc_to_avi_convert([MarshalAs(UnmanagedType.BStr)] String _inputAVC, [MarshalAs(UnmanagedType.BStr)] String _ouputAVI, int pPrivate, PROGRESS_CALLBACK _pg);
[DllImportAttribute(#"..\xxx.dll", EntryPoint = "avc_to_avi_close", CallingConvention = CallingConvention.StdCall, CharSet = CharSet.Unicode)]
public static extern void avc_to_avi_close(ref int avi);
And ran it, and I still get the same error.
1.) Are you sure about the calling convention? Try CallingConvention.StdCall. Read this:
http://blogs.msdn.com/b/adam_nathan/archive/2003/05/21/56690.aspx
2.) Try a different values for the CharSet attribute and use String instead of StringBuilder for the path arguments. This is a good reference:
http://msdn.microsoft.com/en-us/library/s9ts558h.aspx
3.) Also, the avc_to_avi_close method should look like this:
[DllImportAttribute("xxx.dll", EntryPoint="avc_to_avi_close")]
public static extern void avc_to_avi_close(ref int avi);
4.) Other things you my want to try:
Use unusual combinations of CharSet and MarshalAs. For example you know from the link above that Unicode should be used with LPWStr and Ansi with LPStr but you can of course try other combinations.
Your extern methods need not to be marked unsafe.
If you have exhausted all other options, why don't you try to write a C++ program that is analogous to your C# code:
If it crashes you proved that the error is in the library.
If it doesn't crash you know the error is in your use of PInovke, and you can go through all the 12+ combinations from my answer one by one.
You are passing a delegate to p/invoke and then letting the garbage collector free it. This causes the crash, because when the delegate object is finalized, the trampoline set up to allow calls from native code is deallocated. Then the native code invokes a dangling function pointer.
The garbage collector is totally unaware of any objects being used from inside native code. You MUST keep a reference to the delegate alive for the duration of the call. Try adding a delegate variable (don't use implicit conversion which creates a temporary) and then use GCHandle to keep the delegate alive for as long as the native code is using it.
Related
I have a native, unmanaged C++ DLL (symulator.dll) which I have to load in and call from a managed C# application.
The DLL utilizes C++ classes and dynamic memory allocation (via the new operator).
It exports a function called Init and its definition is as follows:
extern "C" __declspec( dllexport ) int Init( void )
{
sym = new CSymulator();
sym->Init();
return 0;
}
The CSymulator class contained within the DLL has a rather simple constructor:
CSymulator::CSymulator( void )
{
memset( mem, 0, sizeof( mem ) );
memset( &rmr, 0, sizeof( rmr ) );
md = 0;
ma = 0;
tacts = 0;
}
The CSymulator::Init() method, called by the Init() function exported by the DLL, is defined as follows:
int CSymulator::Init( void )
{
int *a = new int;
*a = 1;
FILE *f = fopen( "tmp.log", "wb" );
fprintf( f, "%i", *a );
fclose( f );
delete a;
return 0;
}
I am loading the native C++ DLL into the managed C# application using this code:
public partial class Form1 : Form
{
public IntPtr SimHandle;
[DllImport("kernel32.dll", CharSet = CharSet.Auto, SetLastError = true)]
private static extern IntPtr LoadLibrary(string libname);
[DllImport("kernel32.dll", CharSet = CharSet.Auto)]
private static extern bool FreeLibrary(IntPtr hModule);
[DllImport("kernel32.dll", CharSet = CharSet.Ansi)]
private static extern IntPtr GetProcAddress(IntPtr hModule, string lpProcName);
delegate int SimInit();
SimInit DLL_Init;
public void InicjujDLL()
{
IntPtr adres;
adres = GetProcAddress(SimHandle, "Init");
DLL_Init = (SimInit)Marshal.GetDelegateForFunctionPointer(adres, typeof(SimInit));
int rc = DLL_Init();
}
private void WczytajDLL()
{
String fileName = "D:\\prg\\kompilator\\Debug DLL\\symulator.dll";
SimHandle = LoadLibrary(fileName);
if (SimHandle == IntPtr.Zero)
{
int errorCode = Marshal.GetLastWin32Error();
throw new Exception(string.Format("Blad przy wczytywaniu biblioteki ({0})", errorCode));
}
else
{
InicjujDLL();
}
}
private void Form1_Load(object sender, EventArgs e)
{
WczytajDLL();
}
}
This code should produce a file named tmp.log with the content of 1 in it. But for some reason, tmp.log contains garbage data (a random 32-bit integer value instead of 1; for example, 2550276).
It's not the only function that produces garbage output. Any DLL function that tries to allocate memory dynamically is unable to use it after doing so.
It's as if the native C++ DLL is somehow getting its memory purged by the C# garbage collector.
How to prevent this behavior?
Wait a sec: Look at the reference below:
[UnmanagedFunctionPointer(CallingConvention.Cdecl)]
private delegate int MultiplyByTen(int numberToMultiply);
I didn't notice that your delegate doesn't have the same attribute.
For reference, I don't see anything unusual in how you are doing the LoadLibrary:
Dynamically Loading a Native Library
I have done this myself using the exact reference without a problem. I would suggest removing ALL code temporarily that executes inside the DLL and just do a simple pass-through value. Right now Init() is always returning 0. Try something else since you have zeroed out memory before, getting a zero back may just be a side-effect of the mem-zero op. Return 1974 or something.
Ensure you have allow unsafe code enabled and use the memory viewer to look at the stack (you are getting a pointer back so you have a starting point). If you do this beside the IL, you might spot where your memory is getting trashed.
HTH
So I created a custom dll callsed FileGuidUtils.dll written in C/C++ and one of the functions returns a WCHAR * string (as a LPWStr in C#). This string gets allocated heap memory inside the function of type (WCHAR *).
Right now I just use the returned string and that's it. Should I be freeing it somewhere in the C#? What code should I use if so? Or is the CLR garbage collector taking care of it for me?
[DllImport(#"FileGuidUtils.dll", CallingConvention = CallingConvention.Cdecl, CharSet = CharSet.Unicode)]
[return: MarshalAs(UnmanagedType.LPWStr)]
private static extern string getReparseTarget([MarshalAsAttribute(UnmanagedType.LPWStr)] string linkPath);
I occasionally get unhandled out of memory exceptions but am unable to pinpoint the cause because I don't get it very often. For right now I just wanna know if I'm handling the returned string properly?
You are creating a memory leak. .NET can't free memory allocated by malloc, C++ new, or any other allocator of Windows, because it can't know which allocator was used (there are some exceptions on this).
Possible solutions:
1) Return a IntPtr:
private static extern IntPtr getReparseTarget([MarshalAsAttribute(UnmanagedType.LPWStr)] string linkPath);
and have a corresponding
private static extern void freeMemory(IntPtr ptr);
and then manually rebuild the C# string with PtrToStringUni()
2) Have a function that returns the length needed and then pass a StringBuilder of that length (new StringBuilder(len)):
private static extern int getReparseTargetLength([MarshalAsAttribute(UnmanagedType.LPWStr)] string linkPath);
private static extern void getReparseTarget([MarshalAsAttribute(UnmanagedType.LPWStr)] string linkPath, [MarshalAsAttribute(UnmanagedType.LPWStr)] StringBuilder output, int maxLen);
3) Use MarshalAs(UnmanagedType.BSTR)
[return: MarshalAs(UnmanagedType.BSTR)]
private static extern string getReparseTarget([MarshalAsAttribute(UnmanagedType.LPWStr)] string linkPath);
The CLR automatically frees strings allocated as BSTR. Note that you will need to create them as BSTR C++-side (with SysAllocString/SysAllocStringLen).
I have a C++ file with some exported functions which I am calling in C#. One of the functions is this:
char segexpc[MAX_SEG_LEN];
extern "C" QUERYSEGMENTATION_API char* fnsegc2Exported()
{
return segexpc2;
}
Somewhere in the program, I am also doing this thing:
if(cr1==1)
{
strcpy(segexpc, seg);
}
In my C# program, I am invoking the above by the followign way:
[DllImport("QuerySegmentation.dll", EntryPoint = "fnsegcExported", CallingConvention = CallingConvention.Cdecl, CharSet = CharSet.Ansi)]
public static extern StringBuilder fnsegcExported();
this.stringbuildervar = fnsegcExported();
Previously, I was not getting any error, but now suddenly I have started getting this error, when I debug in visual studio.
Windows has triggered a breakpoint in SampleAppGUI.exe.
This may be due to a corruption of the heap, which indicates a bug in SampleAppGUI.exe or any of the DLLs it has loaded.
This may also be due to the user pressing F12 while SampleAppGUI.exe has focus.
This error appears only at the end just before it has to display the window. I have not pressed any F12 key key and neither is any breakpoint set here, but I am not sure why the error is occuring and breaking at this point. this.stringbuildervar = fnsegcExported();
When I press on continue, the window appears with the correct output.
What would happen if you changed your external declaration from
[DllImport("QuerySegmentation.dll", EntryPoint = "fnsegcExported", CallingConvention = CallingConvention.Cdecl, CharSet = CharSet.Ansi)]
public static extern StringBuilder fnsegcExported();
to
[DllImport("QuerySegmentation.dll", EntryPoint = "fnsegcExported", CallingConvention = CallingConvention.Cdecl, CharSet = CharSet.Ansi)]
public static extern string fnsegcExported();
And then called it in the following way:
this.stringbuildervar = new StringBuilder(fnsegcExported());
string seems the more appropriate type. Or better yet use the Marshal class to marshal your unmanaged char* return into a managed string.
[DllImport("QuerySegmentation.dll", EntryPoint = "fnsegcExported", CallingConvention = CallingConvention.Cdecl, CharSet = CharSet.Ansi)]
public static extern IntPtr fnsegcExported();
string managedStr = Marshal.PtrToStringAnsi(fnsegcExported);
this.stringbuildervar = new StringBuilder(managedStr);
The reason why you see the error on this line is that it is the last stack frame for which you have debug information.
Fortunately, in your case, there is very little code on the C++ side. Make sure that segexpc contains a zero terminated string.
I suspect that because the default capacity of the string builder is 16, you may not be able to return longer strings in this fashion. Maybe you want to return just strings.
I also wonder whether your C++ string has to be non-Unicode. That will hurt performance upon every conversion.
I have an unmanaged C++ DLL that I have wrapped with a simple C interface so I can call PInvoke on it from C#. Here is an example method in the C wrapper:
const wchar_t* getMyString()
{
// Assume that someWideString is a std::wstring that will remain
// in memory for the life of the incoming calls.
return someWideString.c_str();
}
Here is my C# DLLImport setup.
[DllImport( "my.dll", CharSet = CharSet.Unicode, CallingConvention = CallingConvention.Cdecl )]
private static extern string GetMyString();
However the string is not correctly marshalled, often screwing up the first character or sometimes way off showing a bunch of chinese characters instead. I have logged output from the implementation on the C side to confirm that the std::wstring is correctly formed.
I have also tried changing the DLLImport to return an IntPtr and convert with a wrapped method using Marshal.PtrToStringUni and it has the same result.
[DllImport( "my.dll", CallingConvention = CallingConvention.Cdecl )]
private static extern IntPtr GetMyString();
public string GetMyStringMarshal()
{
return Marshal.PtrToStringUni( GetMyString() );
}
Any ideas?
Update with Answer
So as mentioned below, this is not really an issue with my bindings but the lifetime of my wchar_t*. My written assumption was wrong, someWideString was in fact being copied during my calls to the rest of the application. Therefore it existed only on the stack and was being let go before my C# code could finish marshalling it.
The correct solution is to either pass a pointer in to my method as described by shf301, or make sure my wchar_t* reference does not get moved / reallocated / destroyed before my C# interface has time to copy it.
Returning the std::wstring down to my C layer as a "const &std::wstring" means my call to c_str() will return a reference that won't be immediately dealloc'd outside the scope of my C method.
The calling C# code then needs to use Marshal.PtrToStringUni() to copy data from the reference into a managed string.
You are going to have to rewrite your getMyString function for the reasons mentioned in Hans Passant's answer.
You need to have the C# code pass a buffer in to your C++ code. That way the your code (ok, the CLR Marshaller) controls the lifetime of the buffer and you don't get into any undefined behavior.
Below is an implementation:
C++
void getMyString(wchar_t *str, int len)
{
wcscpy_s(str, len, someWideString.c_str());
}
C#
[DllImport( "my.dll", CallingConvention = CallingConvention.Cdecl, CharSet = CharSet.Unicode )]
private static extern void GetMyString(StringBuffer str, int len);
public string GetMyStringMarshal()
{
StringBuffer buffer = new StringBuffer(255);
GetMyString(buffer, buffer.Capacity);
return buffer.ToString();
}
You need to specify MarshalAs attribute for the return value:
[DllImport( "my.dll", CharSet = CharSet.Unicode, CallingConvention = CallingConvention.Cdecl)]
[return : MarshalAs(UnmanagedType.LPWStr)]
private static extern string GetMyString();
Make sure the function is indeed cdecl and that the wstring object is not destroyed when the function returns.
I have a problem, I tried to look at almost all the poster solutions, unsuccessful to find a suitable one.
The question is easy, Want to have a return string from unmanaged C code in my managed C#.
The c function is:
extern "C" __declspec(dllexport) int process_batch (char *&result);
and in C# I imported the DLL:
[DllImport("mydll.dll")]
public static extern IntPtr process_batch(StringBuilder result);
I run, but the return value in my StringBuilder is a 7-8 character string of non-sense! (I think the memory address)
I tried adding ref before the StringBuilder, this time I get the correct return value in StringBuilder but I get AccessViolationException :
Attempted to read or write protected memory. This is often an indication that other memory is corrupt.
So I need your help to fix this.
Just one more thing, I use malloc in c for allocating the memory to the char* variable.
Thanks,
If you really want to do that, pass the parameter as a ref IntPtr, then call Marshal.PtrToStringAnsi, or similar.
[DllImport("mydll.dll")]
public static extern IntPtr process_batch(ref IntPtr result);
Note that, since you're allocating the string with malloc in C, the .NET program will have to keep track of the returned pointer so that it can pass it back to you to be deallocated. Otherwise you'll have a memory leak.
If you were to pass a ref StringBuilder, you wouldn't be able to deallocate the memory that was allocated by the C program.
Also, as somebody commented in another post, you need to set the calling convention:
[DllImport("mydll.dll", CallingConvention=CallingConvention.Cdecl)]
I've been using the following code successfully:
[DllImport("user32.dll", EntryPoint = "GetClassName", ExactSpelling = false,
CharSet = CharSet.Auto, SetLastError = true)]
private static extern int _GetClassName(IntPtr hwnd, StringBuilder lpClassName,
int nMaxCount);
public static string GetClassName(IntPtr hWnd)
{
StringBuilder title = new StringBuilder(MAXTITLE);
int titleLength = _GetClassName(hWnd, title, title.Capacity + 1);
title.Length = titleLength;
return title.ToString();
}
I'd advise for a more specific declaration of the imported method via the DllImportAttribute. Try the CharSet = CharSet.Auto bit for instance
I know that this is not exactly related to your original problem as this makes use of the Windows API, but maybe it is of help nonetheless.