[Ktechlab-devel] run crash

[David Saxton]

On Monday 06 February 2006 14:33, Mr Jiff wrote:
> What is supposed to happen when, with the asm file (created from sdcc)
> displayed in the editor, I click on the "Run" icon? All I get is a crash.
>
> The generated cod file works when run by hand (gpsim -s).
>
> Isn't the patched gpsim supposed to solve this crash problem?
>
>
> With the patched gpsim from
> http://ktechlab.org/download/gpsim-0.21.11.tar.bz2 and the ktechlab svn
> from 24 hours ago this is the result
>
> Now with the asm  source tab open and "Run" clicked, here's the trace:
>
> Using host libthread_db library "/lib/tls/libthread_db.so.1".
> `system-supplied DSO at 0xffffe000' has disappeared; keeping its symbols.
> [Thread debugging using libthread_db enabled]
> [New Thread 1110000224 (LWP 2354)]
> [KCrash handler]
> #4  0x00000000 in ?? ()
> #5  0x405a7903 in Symbol_Table::clear (this=0x4062b8c0) at symbol.cc:739
> #6  0x40599bbc in ProgramFileTypeList::LoadProgramFile (this=0x8430050,
>     pProcessor=0xbf84dd44,
>     pFilename=0x8634bf0 "/tmp/kde-gh/ktechlabmo4jkb.cod", pFile=0x87134f8)
>     at program_files.cc:98

This backtrace looks identical to the crash caused by the bug that the patch 
fixes. So yes, the patch should fix it.....

Just to be sure, could you check you *definitely* are using the patched gpsim 
version (and haven't e.g. simply installed the patched gpsim to the wrong 
location). By the nature of the Symbol_Table::clear function (the one in the 
backtrace), it seems very likely that this is being caused by the bug that 
the patch addresses.

If you still experience the crash, then you could try sending me the file that 
it crashes on (unless it crashes on any file)...

David