This is the mail archive of the gcc-help@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

unknown bound: array error.



I have a large amount of legacy code for a project that we support
on 
various unix platforms. Some of the internal structures are defined
as:

typedef struct {
 int xpos;
 int ypos;
} XYCoordStruct, (*XYCoordPtr)[];

Note the pointer/array declaration. My belief is that at some point
in 
Ancient times someone scattered those about to solve some array 
problems they may have had with a ** declared variable.

This has never caused us a problem until I tried compiling with GCC
3.0 
on Linux and Solaris 8. Gcc is complaining about function calls
using the 
variables.  Worst thing is, its treating this as an error, not a
warning. 
ouch.

/diska/INCLUDE/libData.h:236: parameter `rpts' includes
   pointer to array of unknown bound `XYCoordStruct[]

where "libData.h" has a call like this:

int  libdReturnVCoords (
    libfHandle, int id_index, int time, int npts,
    XYCoordPtr rpts, int *nearest );

I think that I've fairly exhaustively examined all the documentation 
concerning command line switches, and any that seem like they might 
help, haven't.

Does anyone know of a switch that will allow gcc to let these
through 
without complaint?   There is too much code (that works fine with
all the 
other compilers we use) to change, so that isn't an option.

Thanks.


-- 
-MJT
[==========================================]
  Mark J. Turick             
  mailto:mturick@mrburns npt nuwc navy mil 
[==========================================]


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]