[Pkg-utopia-maintainers] Bug#941703: [Pkg-gtkpod-devel] Bug#941703: libimobiledevice6: Crashes upower with stack smashing when connecting an iPhone

[Diego Escalante Urrelo]

Hi

While trying to play around with this, I rebuild libimobiledevice
locally from the sources available in unstable and the crash went
away.
If I reinstall the .deb from the repositories I can get the crash
back, reinstalling the locally built package fixes the problem (from
apt source + dpkg-buildpackage -us -uc).

I guess this is still a libimobiledevice bug since I can reproduce the
stack smash by running ideviceinfo and a few other of the idevice
utilities (idevicename, idevicesyslog):
(gdb) r
Starting program: /usr/bin/ideviceinfo
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
*** stack smashing detected ***: <unknown> terminated

Program received signal SIGABRT, Aborted.
__GI_raise (sig=sig at entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
50    ../sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt
#0  __GI_raise (sig=sig at entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
#1  0x00007ffff7dd0535 in __GI_abort () at abort.c:79
#2  0x00007ffff7e26db8 in __libc_message (action=<optimized out>,
fmt=fmt at entry=0x7ffff7f318a2 "*** %s ***: %s terminated\n")
    at ../sysdeps/posix/libc_fatal.c:181
#3  0x00007ffff7eb581d in __GI___fortify_fail_abort
(need_backtrace=need_backtrace at entry=false,
msg=msg at entry=0x7ffff7f31880 "stack smashing detected")
    at fortify_fail.c:28
#4  0x00007ffff7eb57d2 in __stack_chk_fail () at stack_chk_fail.c:29
#5  0x00007ffff7f744b7 in idevice_new () from
/lib/x86_64-linux-gnu/libimobiledevice.so.6
#6  0x0000000000000000 in ?? ()

On Fri, Oct 4, 2019 at 2:26 AM Yves-Alexis Perez <corsac at debian.org> wrote:
>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
> control: reassign -1 upower
>
> On Thu, 2019-10-03 at 18:23 -0500, Diego Escalante Urrelo wrote:
> > Whenever you connect an iPhone when upower is running, a crash in upower
> > is triggered, apparently because libimobiledevice is doing something
> > leading to a stack smash crash.
> >
> > The same happens if you already have the iPhone connected when upower
> > starts. I'm attaching a trace and log of the first case (connecting the
> > iPhone when upower is already running).
> >
> > Note that this crash triggers upower to endlessly reload because of the
> > crash-restart-crash cycle it gets into.
>
> Hi Diego,
>
> that looks spurious indeed, and I don't think I ever experienced that. When
> did this behavior appeared and can you link it to a specific upgrade?
>
> Also, I'm really not sure it's a problem in libimobiledevice. If UPower
> crashes, then it's likely a problem in UPower (maybe because it doesn't like
> the way a new battery is appearing or something).
>
> I'm reassigning to UPower and add their maintainer to CC:.
>
> @UPower maintainer: feel free to reassign back if you identify something wrong
> in libimobiledevice.
>
> Regards,
> - --
> Yves-Alexis
> -----BEGIN PGP SIGNATURE-----
>
> iQEzBAEBCAAdFiEE8vi34Qgfo83x35gF3rYcyPpXRFsFAl2W8/kACgkQ3rYcyPpX
> RFt2KQf9ETUPRbi30PtbN/dEwIZFDZwxZt8tGP/hw7ewn4Iq5ctQed5O2uyknC41
> txfBr2AuwnfIOcqr/0swRNRJ2t7pRvX6XJL3MmY11KLtI7FtZ/RJ4CdZmRLQGYRO
> b3rso4ZP7ueQcTpIYr1Fy5nXPBM/Mc4m8wKzFklYu6lc4izYDDCIoUJknBMudn2S
> YxUc72zSln6r3ExrPsr/XFnfMNH/gGTlKSmsyfcK7wQJk9B5e4Ofjvc49f+hU/gK
> 1yzvHnycwDz2QPahS5NNB0S7n/TmTiIXZ9rJLL0jQsFBYhu4FuShxY+5boRZvxaF
> 1tQsnWmTHxkhhrZ+wbWJVnS45Eu0TA==
> =pkib
> -----END PGP SIGNATURE-----