option driver freezing entire system
Kasper Sandberg
Im having some issues with some huawei e1752 modems, the computer
sometimes completely freezes, and i suspect its due to the option
driver, as it doesnt happen if i dont modeswitch them to appear as
option modem.
Do you have any idea what might cause this? i looked all over the system
logs and theres nothing to help at all, but i have seen this spanning
from .26 to .32, on multiple computers, and i just dont know what to do
to fix it.
Thanks in advance
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Oliver Neukum
> Hello..
>
> Im having some issues with some huawei e1752 modems, the computer
> sometimes completely freezes, and i suspect its due to the option
> driver, as it doesnt happen if i dont modeswitch them to appear as
> option modem.
>
> Do you have any idea what might cause this? i looked all over the system
> logs and theres nothing to help at all, but i have seen this spanning
> from .26 to .32, on multiple computers, and i just dont know what to do
> to fix it.
Please recompile your kernel with lockdep. This should improve
debugging.
Regards
Oliver
Kasper Sandberg
> Am Freitag, 15. Januar 2010 18:38:36 schrieb Kasper Sandberg:
> > Hello..
> >
> > Im having some issues with some huawei e1752 modems, the computer
> > sometimes completely freezes, and i suspect its due to the option
> > driver, as it doesnt happen if i dont modeswitch them to appear as
> > option modem.
> >
> > Do you have any idea what might cause this? i looked all over the system
> > logs and theres nothing to help at all, but i have seen this spanning
> > from .26 to .32, on multiple computers, and i just dont know what to do
> > to fix it.
>
> Please recompile your kernel with lockdep. This should improve
> debugging.
It is a debian unstable .32 kernel i run now, and grepping for lockdep
reveals:
CONFIG_LOCKDEP_SUPPORT=y
only, is there anything more i should have? (never used lockdep before)
Jiri Kosina
> > Please recompile your kernel with lockdep. This should improve
> > debugging.
>
> It is a debian unstable .32 kernel i run now, and grepping for lockdep
> reveals:
> CONFIG_LOCKDEP_SUPPORT=y
> only, is there anything more i should have? (never used lockdep before)
CONFIG_PROVE_LOCKING could be helpful here. (be aware of the fact that
this will make your system noticeably slower).
--
Jiri Kosina
SUSE Labs, Novell Inc.
Kasper Sandberg
> On Fri, 15 Jan 2010, Kasper Sandberg wrote:
>
> > > Please recompile your kernel with lockdep. This should improve
> > > debugging.
> >
> > It is a debian unstable .32 kernel i run now, and grepping for lockdep
> > reveals:
> > CONFIG_LOCKDEP_SUPPORT=y
> > only, is there anything more i should have? (never used lockdep before)
>
> CONFIG_PROVE_LOCKING could be helpful here. (be aware of the fact that
> this will make your system noticeably slower).
Sorry, i was just away a few days. I shall enable this setting tomorow
and test, though it may take a few days for it to crash.
Jiri Kosina
> > > > Please recompile your kernel with lockdep. This should improve
> > > > debugging.
> > >
> > > It is a debian unstable .32 kernel i run now, and grepping for lockdep
> > > reveals:
> > > CONFIG_LOCKDEP_SUPPORT=y
> > > only, is there anything more i should have? (never used lockdep before)
> >
> > CONFIG_PROVE_LOCKING could be helpful here. (be aware of the fact that
> > this will make your system noticeably slower).
>
> Sorry, i was just away a few days. I shall enable this setting tomorow
> and test, though it may take a few days for it to crash.
This option should ideally report the lock ordering violation before the
actual hang happens (if it is caused by improper locking semantics).
--
Jiri Kosina
SUSE Labs, Novell Inc.
--