gdb slate on sparc

[Nick Forde]

Hi Lee,

Will the slate VM run on a 64-bit machine with the alpha big.image?
Although this is a similar stack dump to the one discussed back in
March (http://lists.tunes.org/archives/slate/2005-March/001170.html)
I don't think it is due to the GCC UltraSPARC struct alignment  
optimization.
Helmut mentioned on IRC that switching to -O0 doesn't help.

Nick.

On 7 Sep 2005, at 21:40, Helmut Grohne wrote:

> Hi,
>
> I investigated the boostrap bug on sparc on a sparc64. Optimization  
> was
> switched off. Instead -O0 and -ggdb was added to CFLAGS. Here are my
> results:
>
> $ uname -a
> Linux ypsilon 2.4.27-1-sparc64 #1 Mon Aug 23 23:59:55 PDT 2004 sparc64
> NU/Linux
> $ ./debian/endianess
> big
> $ gdb ./vm
> ...
> (gdb) run big.image
> Starting program: /public_home/debian/helmutg/slate-0.3.5/vm big.image
>
> Program received signal SIGBUS, Bus error.
> 0x0001cbd8 in ObjectPointer_dispatchTo_arity_above_ (name=1881500452,
>     arguments=0x70296d84, n=1, resendMethod=0) at ../../../../../ 
> vm.c:3181
> 3181                if (!(map -> dispatchID == currentDispatchID))
> (gdb) display map
> 1: map = (struct Map *) 0x7029f6e4
> (gdb) display *map
> 2: *map = {header = {isMarked = 1, idHash = 15400, objectSize = 8,
>     objectFormat = 2}, map = 0x702a2740, flags = 1,
>   representative = 1881206236, delegates = 0x7029f714, slotCount = 1,
>   slotTable = 0x7029f72c, roleTable = 0x7029f758, dispatchID = 0,
>   visitedPositions = 0}
> (gdb)
>
> Helmut
>