WSARecv

[Rage9]

Been working on this problem for more hours than I'd like to admit.

It seems to me it will be an easy-fix but my brain is fried over this issue.  :dazzled:

WSARecv keeps failing and WSAGetLastError keeps returning:

WSAEFAULT
10014    

Bad address.
    The system detected an invalid pointer address in attempting to use a pointer argument of a call. This error occurs if an application passes an invalid pointer value, or if the length of the buffer is too small. For instance, if the length of an argument, which is a sockaddr structure, is smaller than the sizeof(sockaddr).

Code Select Expand


WSABUF STRUCT
  len    DWORD   ?
  buf   DWORD   ?
WSABUF ENDS

AcceptConnection Proc uses ebx

       local   dwFlags:DWORD,
dwBytes:DWORD,
                wbuf:WSABUF,
ol:OVERLAPPED

;associating with IOCP stuff

                mov dwFlags, 0
        mov dwBytes, 0

                invoke WSARecv, [ebx].hSocket, addr wbuf, 1,  addr dwBytes,  dwFlags,  addr ol, NULL




Return Values

If no error occurs and the receive operation has completed immediately, WSARecv returns zero. Note that in this case, the completion routine will have already been scheduled, and to be called once the calling thread is in the alertable state. Otherwise, a value of SOCKET_ERROR is returned, and a specific error code may be retrieved by calling WSAGetLastError. The error code WSA_IO_PENDING indicates that the overlapped operation has been successfully initiated and that completion will be indicated at a later time. Any other error code indicates that the overlapped operation was not successfully initiated and no completion indication will occur.

In this case WSAEFAULT.  Driving me nuts - thanks in advance.

[Tedd]

Simple check first: what's the value of ebx?

(Also, I'd give NULL for lpOverlapped if you're not using it; if you are then lpCompletion shouldn't be null.)

[japheth]

at least check how OVERLAPPED has to be initialized. It isn't allowed to hold random values.

[Rage9]

It has been pointed out to me that my problem lies in the fact that the structure is defined incorrectly.

Ex found here: http://msdn2.microsoft.com/en-us/library/ms741542.aspx

Most noticeably seen:

Code Select Expand


WSABUF DataBuf;
char SendBuf[1024] = "Test data to send.";
int BufLen = 1024;

...

DataBuf.len = BufLen;
DataBuf.buf = SendBuf;


As seen .buf is an actual array hence of characters and .len is the actual length of the buffer.

I change

Code Select Expand


WSABUF STRUCT
  len    DWORD  ?
  buf   DWORD   ?
WSABUF ENDS


To

Code Select Expand


WSABUF STRUCT
  len    db 256 DUP(?)
  buf   DWORD   ?
WSABUF ENDS


Now when trying to compile it, the compiler process just hangs.

Thoughts?

[PBrennick]

Probably because you are overwriting other parts of data. I think it should be:

Code Select Expand


WSABUF STRUCT
  buf    db 256 DUP(?)
  len   DWORD   ?
WSABUF ENDS


Paul

[MichaelW]

Rage9,

Code Select Expand


typedef struct _WSABUF {
    u_long      len;     /* the length of the buffer */
    char FAR *  buf;     /* the pointer to the buffer */
} WSABUF, FAR * LPWSABUF;


A "u_long" is a DWORD and a "char FAR *" is a DWORD.

[edike123]

Rage 9,

I had the same problem, it can also occur if setsockopt is not invoked on [ebx].hSocket.

check out the following post: http://www.masm32.com/board/index.php?topic=15547.0

[dedndave]

change this...

Code Select Expand

WSABUF STRUCT
  len   db 256 DUP(?)
  buf   DWORD   ?
WSABUF ENDS
to this...

Code Select Expand

WSABUF STRUCT
  len   DWORD ?
  buf   db 256 DUP(?)
WSABUF ENDS
or...

Code Select Expand

WSABUF STRUCT
  len   dd ?
  buf   db 256 DUP(?)
WSABUF ENDS

edlike.....
this thread is 3 years old   :P

[edike123]

edlike.....
this thread is 3 years old   :P

dedndave,

I know that but I still think it is worth responding, because googling  "WSARecv error" easily redirects to this post, so maybe others having similar problems will be able to find solutions more efficiently. (sorry for being off topic)

[dedndave]

no problem   :P

sometimes, people post without realizing the thread is old