actually problem starts from 4.53.

1. build and run 4.52 is fine

2. 4.53 crashes

3. setup gdb, as long as i connect to target, it says cannot access memory 0.

after i set breakpoint at main() in stunnel.c, it get seg fault signal when i hit "continue" in gdb.

panic ...

From: ningji@hotmail.com
To: stunnel-users@stunnel.org
Date: Sat, 15 Dec 2012 15:43:09 +0000
Subject: [stunnel-users] core dump when running stunnel on Arm based board

Hi everyone, i'm new to this.

got a task to update stunnel from 4.52 to 5.00b1.

1. ./configure --host arm-none-linux-gnueabi --target arm-none-linux-gnueabi --prefix=

i compared the makefile with the one in our 4.52 code base, same.

2. build with

make -C stunnel CROSS_COMPILE=arm-none-linux-gnueabi- ARCH=arm

3. scp to ARM board, run without arugments, core dump.

(the line number might be different as i was putting printfs. but as long as it enters parse_commandline, it crashes)

[New LWP 867]

Cannot access memory at address 0x70797263
(gdb) bt
#0 parse_commandline (name=0xbefdce29 "/opt/ultra/config/stunnel.conf",
parameter=0xbefdce48 "-version") at options.c:1940
#1 0x0003ba04 in main_configure (
arg1=0xbefdce29 "/opt/ultra/config/stunnel.conf",
arg2=0xbefdce48 "-version") at stunnel.c:199
#2 0x0003bd44 in main_unix (argv=<optimized out>, argc=<optimized out>)
at stunnel.c:135
#3 main (argc=3, argv=0xbefdcd24) at stunnel.c:110

my question is, it crashed when just entering parse_commandline( ),
the 2 arguments look fine to me.
How could this happen ?

Many thanks !

_______________________________________________ stunnel-users mailing list stunnel-users@stunnel.org https://www.stunnel.org/cgi-bin/mailman/listinfo/stunnel-users