Re: Dawrin- why does pinky SIGBUS

[mwoehlke]

(Sigh. I sure wish gdb built (and worked) on Darwin, Apple's version 
insists on invoking a broken csh that bitches about my modern LS_COLORS.)

Paul Eggert wrote:
> mwoehlke <address@hidden> writes:
>
> > #0  0x90007260 in strlen ()
> > #1  0x9000d1d8 in strdup ()
> > #2  0x0000393c in canon_host_r (host=0xfffffffc <Address 0xfffffffc
> > out of bounds>, cherror=0x8248) at canon-host.c:74
>
> This suggest that this code in canon-host.c:
>
>   hints.ai_flags = AI_CANONNAME;
>   status = getaddrinfo (host, NULL, &hints, &res);
>   if (!status)
>     {
>       retval = strdup (res->ai_canonname);
>
> set status to 0, but res->ai_canonname was NULL.  Can you verify
> that this in fact happened?

(gdb)
69        status = getaddrinfo (host, NULL, &hints, &res);
(gdb) step
70        if (!status)
(gdb) print status
$1 = 0
(gdb) print *res
$2 = {
  ai_flags = 0,
  ai_family = 2,
  ai_socktype = 2,
  ai_protocol = 17,
  ai_addrlen = 16,
  ai_canonname = 0x0,
  ai_addr = 0x30b720,
  ai_next = 0x3009e0
}
(gdb)

Yup, that's what happened. getaddrinfo() broken?

> Are you using the system getaddrinfo, or the one shipped with
> coreutils?

Um... no idea? I un-tar'd, did './configure --prefix=<dir on an NFS 
mount>' and 'make'. So I'm using whatever that picks up? If it means 
anything, when I 'step'd in gdb at the getaddrinfo call, it stepped over 
rather than into (not that I entirely trust gdb 5.3 with Apple patches).

That said, given the address when I set a breakpoint there, I am 
guessing it is the system getaddrinfo?
(gdb) break getaddrinfo
Breakpoint 1 at 0x9005492c
(gdb) run
Starting program: /home/install/gnu/alpha/src/coreutils-6.3/src/pinky
Reading symbols for shared libraries +.. done
Reading symbols for shared libraries . done
[snip]
Reading symbols for shared libraries . done
Login    Name                 TTY      Idle   When         Where

Breakpoint 1, 0x9005492c in getaddrinfo ()
(gdb) bt
#0  0x9005492c in getaddrinfo ()
#1  0x00002bd8 in print_entry (utmp_ent=0xbffff8c0) at pinky.c:305
#2  0x00002bd8 in print_entry (utmp_ent=0x18013b0) at pinky.c:305
#3  0x0000320c in scan_entries (n=2, utmp_buf=0x18013b0, argc_names=0, 
argv_names=0xbffffb08) at pinky.c:476
#4  0x000032a4 in short_pinky (filename=0x796c "/var/run/utmp", 
argc_names=25170864, argv_names=0xbffffb08) at pinky.c:494
#5  0x00003684 in main (argc=1, argv=0xbffffb04) at pinky.c:622
(gdb)

> > Honestly I have never used pinky so I don't really care.
>
> It's a sign of problems in the library, though.  It should
> get fixed.

Clearly you talked me into it. :-)

--
Matthew
The hippo made me do it! What? What do you mean you can't see the hippo?