Re: BUG: All network processes hang (brcmsmac/wpa_supplicant)

[Eric Dumazet <eric.dumazet@gmail.com>]

Le mercredi 19 octobre 2011 à 14:58 +0200, Arend van Spriel a écrit :
> On 10/19/2011 02:49 PM, Eric Dumazet wrote:
> > Le mercredi 19 octobre 2011 à 14:36 +0200, Nico Schottelius a écrit :
> >> Hello,
> >>
> >> I've
> >>
> >> Eric Dumazet [Tue, Oct 18, 2011 at 02:24:25PM +0200]:
> >>> Must be a mutex_unlock(some_mltex) missing somewhere.
> >>>
> >>> Then later, a process holding RTNL is blocking on mutex_lock(some_mutex)
> >>>
> >>> Try a "CONFIG_LOCKDEP=y" enabled build
> >>
> >> Did so, attached is the dmesg output after 3x suspend.
> >>
> >> It also seems this bug gets more / only triggered when I loose the
> >> connection to a wpa2-eap network that is defined like this:
> >>
> >> network={
> >>    ssid="eth"
> >>   id_str="eth-802.2x"
> >>    key_mgmt=WPA-EAP
> >>    proto=WPA2
> >>    eap=TTLS
> >>    pairwise=CCMP TKIP
> >>    group=CCMP TKIP
> >>    phase2="auth=MSCHAPV2"
> >>    identity="MYNAME"
> >>    password="forgottoremoveit"
> >> }
> >>
> >> Hope the dmesg output helps and we can remove the problem soon
> >> and making the mba finally almost completly usable :-)
> >>
> > 
> > Sorry, no CONFIG_LOCKDEP in your config or dmesg. Maybe you sent a
> > previous dmesg/config file ?
> > 
> > Note that you have "CONFIG_LOCKDEP_SUPPORT=y", not "CONFIG_LOCKDEP=y"
> > 
> 
> Hi Eric,
> 
> Are there any configuration setting needed that CONFIG_LOCKDEP depends on?



config LOCKDEP
        bool
        depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
        select STACKTRACE
        select FRAME_POINTER if !MIPS && !PPC && !ARM_UNWIND && !S390 && !MICROBLAZE
        select KALLSYMS
        select KALLSYMS_ALL

make menuconfig
 -> Kernel hacking
   -> Lock debugging: prove locking correctness