Re: "Too many arguments", SuSE 6.2

[Matt Aylward]

On Mon, 29 Nov 1999, you wrote:
Marcus G. Daniels: >>>>>> "MA" == Matt Aylward <address@hidden> writes:
Marcus G. Daniels: >
Marcus G. Daniels: >MA> This suggests that it is bailing out during a recursive
Marcus G. Daniels: >MA> neighbourhood search to determine the size and stats of 
contiguous
Marcus G. Daniels: >MA> seagrass patches for agent classification purposes. I 
assume
Marcus G. Daniels: >MA> something has changed in swarm or the objc which is 
stricter on
Marcus G. Daniels: >MA> recursion. 
Marcus G. Daniels: >
Marcus G. Daniels: >I don't know what that would be.  What is the backtrace at 
the point
Marcus G. Daniels: >of the segfault?  What do you get from:

The backtrace goes on for ever with the search (localSearchCell), for
instance:

(gdb) bt 10
#0  objc_msg_lookup (receiver=0x8153230, op=0x805d950)
    at /home/matt/src/swarm/swarm-1999-11-21/libobjc/sendmsg.c:143
#1  0x8055279 in _i_PatchWork__localSearchCellX_Y_Input_ (self=0x8153230, 
    _cmd=0x805d950, x=26, y=92, input=0x81f7108) at PatchWork.m:190
#2  0x8055279 in _i_PatchWork__localSearchCellX_Y_Input_ (self=0x8153230, 
    _cmd=0x805d950, x=26, y=91, input=0x81f7108) at PatchWork.m:190
#3  0x8055279 in _i_PatchWork__localSearchCellX_Y_Input_ (self=0x8153230, 
    _cmd=0x805d950, x=27, y=92, input=0x81f7108) at PatchWork.m:190
#4  0x8055279 in _i_PatchWork__localSearchCellX_Y_Input_ (self=0x8153230, 
    _cmd=0x805d950, x=27, y=93, input=0x81f7108) at PatchWork.m:190
#5  0x8055279 in _i_PatchWork__localSearchCellX_Y_Input_ (self=0x8153230, 
    _cmd=0x805d950, x=27, y=94, input=0x81f7108) at PatchWork.m:190
#6  0x8055279 in _i_PatchWork__localSearchCellX_Y_Input_ (self=0x8153230, 
    _cmd=0x805d950, x=27, y=95, input=0x81f7108) at PatchWork.m:190
#7  0x8055279 in _i_PatchWork__localSearchCellX_Y_Input_ (self=0x8153230, 
    _cmd=0x805d950, x=27, y=96, input=0x81f7108) at PatchWork.m:190
#8  0x8055279 in _i_PatchWork__localSearchCellX_Y_Input_ (self=0x8153230, 
    _cmd=0x805d950, x=27, y=97, input=0x81f7108) at PatchWork.m:190
#9  0x8055279 in _i_PatchWork__localSearchCellX_Y_Input_ (self=0x8153230, 
    _cmd=0x805d950, x=27, y=98, input=0x81f7108) at PatchWork.m:190
(More stack frames follow...)
(gdb)

The localSearchCell method code in PatchWork.m is as follows:
This method is in the "PatchWork" class which is a SwarmObject


-localSearchCellX: (int) x Y: (int) y Input: (id <Discrete2d>) input
{
  int xLocn, yLocn;

  for (xLocn = x + 1; xLocn >= x - 1; xLocn--) {
    for (yLocn = y - 1; yLocn <= y + 1; yLocn++) {
      if (xLocn >= 0 && xLocn < xSize && yLocn >= 0 && yLocn < ySize) {
        if ((int) *discrete2dSiteAt (lattice, offsets, xLocn, yLocn) == 1) {
          if (++area < minArea) {
            cells[area][0] = xLocn;
            cells[area][1] = yLocn;
          }
        [input putValue: CORE atX: xLocn Y:yLocn];
            coreCount++;
        [self localSearchCellX: xLocn Y: yLocnInput:input];  --------LINE #190
        }
      }
    }
  }
  return self;
}



Marcus G. Daniels: >
Marcus G. Daniels: >(gdb) print **(struct objc_class **)receiver
Marcus G. Daniels: >(gdb) print *op
Marcus G. Daniels: >(gdb) call sel_get_name (op)

Program received signal SIGSEGV, Segmentation fault.
objc_msg_lookup (receiver=0x8153230, op=0x805d950)
    at /home/matt/src/swarm/swarm-1999-11-21/libobjc/sendmsg.c:143
143       if(receiver)
(gdb) print **(struct objc_class **)receiver
$1 = {class_pointer = 0x805d840, super_class = 0x40152f60, 
  name = 0x805919a "PatchWork", version = 0, info = 16646157, 
  instance_size = 2108, ivars = 0x805d6e0, methods = 0x805d7a0, 
  dtable = 0x8135ca0, subclass_list = 0x0, sibling_class = 0x400cfcc0, 
  protocols = 0x0, gc_object_type = 0x0}
(gdb) print *op
$2 = {sel_id = 0x90026, sel_types = 0x80593dc "@address@hidden:address@hidden"}
(gdb) call sel_get_name (op)
Cannot access memory at address 0xbfe01ffc
(gdb) 

I think it is time for me to re-read the gdb manual, it has been a while.
Thanks,
Matt

--
      Matt Aylward <address@hidden>
Deptartment of Botany, University of Western Australia
   Long live Linux, Swarm modelling and Grass GIS.

                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.