[PD] splitfilename bug?

[Johannes Burström]

Hi

IOhannes m zmoelnig wrote:
 > unfortunately i cannot really reproduce this.
 > how many symbols do you consider "a few" ?

"a few" would be anything from 2-3 to 50 (sometimes even more - and very 
randomly). Thats "fdshiol<enter>fjsdkl<enter>...[repeat until crash]". 
And here are my outputs:

pd -stderr:
free(): invalid pointer 0x8215550!
pd_gui: pd process exited
Segmenteringsfel

gdb:

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 16384 (LWP 1115)]
0x401ece03 in mallopt () from /lib/libc.so.6
(gdb) bt
#0  0x401ece03 in mallopt () from /lib/libc.so.6
#1  0x401ec763 in calloc () from /lib/libc.so.6
#2  0x0809d935 in getbytes (nbytes=136562160) at m_memory.c:24
#3  0x0809da84 in binbuf_text (x=0x8213fd8,
text=0xbfffd680 "pd ping ; key 1 13 0;;;_L 0 ; 1 ;", size=9)
at m_binbuf.c:70
#4  0x080a8905 in socketreceiver_doread (x=0x8213ff8) at s_inter.c:402
#5  0x080a6ad8 in socketreceiver_read (x=0x8213ff8, fd=555) at s_inter.c:505
#6  0x080a86cf in sys_domicrosleep (microsec=136562160, pollem=1)
at s_inter.c:170
#7  0x080a7db8 in sys_microsleep (microsec=136562160) at s_inter.c:184
#8  0x080a340f in m_scheduler () at m_sched.c:481
#9  0x080ad23b in main (argc=136562160, argv=0x823c5f0) at s_entry.c:27

Thanks,
Johannes


 >
 > but you could run a debugger to see where it crashes:
 > $ gdb pd
 > % run -lib iemlib2 splitfilenametest.pd
 > % bt
 >
 > and post the output of "bt".
 >
 > additionally it often helps to run pd with "-stderr" so any 
error-messages will not immediatly vanish with the tk-console when pd dies.
 >
 >
 >
 > mfg.a.er
 > IOhannes
 >
 > _______________________________________________
 > PD-list at iem.at mailing list
 > UNSUBSCRIBE and account-management -> 
http://lists.puredata.info/listinfo/pd-list
 >
 >
 >