[lug] Netscape Crash Trace

[Chip Atkinson]

It looks like the last thing it did was a brk, which is what a malloc or 
realloc becomes at the system level I believe.  Given that, it may be 
that they are doing things to mess up previously allocated memory, such 
as walking off the end of an array or something.  I'd go with D.Stimits' 
suggestion. :-}

Kyle Moore wrote:

> No...I got tons of stuff. I just didn't know how much of it was relevant
> to the problem. Also, it doesn't cause a problem on Mozilla or Netscape
> 6.0 on the same box. I have also tested with different versions of
> Netscape and IE on different versions of Windows and had no problems.
> 
> This is a bit more of what I got. Even when I wasn't doing anything,
> Netscape kept the xterm (with strace in it) scolling.
> 
> 
> --- SIGALRM (Alarm clock) ---
> gettimeofday({1001450338, 35609}, NULL) = 0
> sigreturn()                             = ? (mask now [])
> time(NULL)                              = 1001450338
> gettimeofday({1001450338, 46237}, NULL) = 0
> time(NULL)                              = 1001450338
> write(12, "\372", 1)                    = 1
> time(NULL)                              = 1001450338
> write(8, "\2\0\4\0\206\4\0\5\2\0\0\0\377\377\377\0F\0\7\0F\4\0\5"...,
> 2040) = 2040
> gettimeofday({1001450338, 46949}, NULL) = 0
> time(NULL)                              = 1001450338
> write(12, "\372", 1)                    = 1
> rt_sigprocmask(SIG_BLOCK, NULL, [], 8)  = 0
> gettimeofday({1001450338, 47246}, NULL) = 0
> rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
> rt_sigprocmask(SIG_BLOCK, NULL, [], 8)  = 0
> gettimeofday({1001450338, 47402}, NULL) = 0
> rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
> brk(0x92fe000)                          = 0x92fe000
> --- SIGSEGV (Segmentation fault) ---
> getpid()                                = 21648
> kill(21648, SIGBUS)                     = 0
> --- SIGBUS (Bus error) ---
> +++ killed by SIGBUS +++
> 
> 
> _______________________________________________
> Web Page:  http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug