Re: malloc errors and SwarmFest

[glen e. p. ropella]

> can you tell me what generally is the problem if I receive a seg
> fault in _smalloc ? My guess is that I'm overwriting a used piece of
> memory - is that right - I'm getting this infuriatingly intermittent
> bug at the following stage when I'm creating an object instance in
> findYoungHeatherAround_Y_ (the create is done using globalZone)....
> 
> #0  0xef5bb850 in _smalloc ()
> #1  0xef5bb884 in malloc ()
> #2  0x7946c in _i_Zone_c__allocIVars_ (self=0x18a5d0, _cmd=0x152164, 
>     aClass=0x13dfdc) at Zone.m:97
> #3  0x78c8c in _c_CreateDrop_s__createBegin_ (self=0x13dfdc, _cmd=0x152154, 
>     aZone=0x18a5d0) at Create.m:38
> #4  0x78c04 in _c_CreateDrop_s__create_ (self=0x13dfdc, _cmd=0x13e04c, 
>     aZone=0x18a5d0) at Create.m:28
> #5  0x26004 in _i_Grouse__findYoungHeatherAround_Y_ (self=0x2a8ab8, 
>     _cmd=0x140770, px=0xeffff00c, py=0xeffff008) at Heatbug.m:418
> #6  0x2e5d8 in _i_ForageExpert1__forage_ (self=0x20c1f0, _cmd=0x14299c, 
>     heatbuglist=0x260a80) at forage.m:258
> #7  0x7625c in _i_Object_s__perform_with_ (self=0x20c1f0, _cmd=0x14bb3c, 
>     aSel=0x14299c, anObject1=0x260a80) at DefObject.m:250
> #8  0x5ea30 in _i_ActionTo_1___performAction__ (self=0x2782a8, _cmd=0x14bf6c, 
>     anActivity=0x288a30) at Action.m:191
> #9  0x60588 in _i_Activity_c___run_ (self=0x288a30, _cmd=0x14bf54)
>     at Activity.m:482

This doesn't really help that much.  If you could post some
source code and do a little context switching in gdb after
it bombs (using "up" a few times works quite well), maybe 
I could help.  But, I recently tried to help Sven that way
and failed utterly. [grin]

> p.s. is there still room for one more at SwarmFest - I'd love to
> come and find out more about what else is going on in the Swarm
> community *and*, if you think people would be interested, I can give
> a demo of a workbench I'm developing usingthe Swarm toolkit that
> facilitates the setting up and comparison of alternative domain
> simulations based on conflicting expert theories. My example domain
> is the politically thorny and often bloody avian world of red grouse
> population dynamics and their management in Scotland. The blood
> flowing more often between arguing scientists than between the birds
> themselves ...  I'll have a web site sorted out by Christmas to post
> on the user community page if you want more info!

Definitely!!  At present, we seem to only have an attendance of 
about 15, so we need to troll for dates again.  But, all demos are
welcome and I'm sure a talk or presentation would be also.

glen