Re: [lockdep warning] INFO: inconsistent lock state, serial8250_interrupt(), &port_lock_key

[Ingo Molnar <mingo@elte.hu>]

* Borislav Petkov <petkovbb@googlemail.com> wrote:

> On Thu, Jul 24, 2008 at 08:53:29AM +0200, Borislav Petkov wrote:
> > On Wed, Jul 23, 2008 at 11:36:04AM +0200, Ingo Molnar wrote:
> > > 
> > > * Ingo Molnar <mingo@elte.hu> wrote:
> > > 
> > > > =================================
> > > > [ INFO: inconsistent lock state ]
> > > > 2.6.26-tip-06509-gb4ebc67-dirty #13600
> > > > ---------------------------------
> > > 
> > > the upstream component of that is: v2.6.26-6077-gc010b2f
> > > 
> > > i.e. my suspicion is that this got introduced via the recent tty 
> > > changes.
> > 
> > Hi,
> > 
> > i hit the same warning here. How about the following fix (this is at least what
> > i think happens):
> > --
> > 
> > serial8250_startup() might unconditionally enable irqs after releasing
> > &up->port.lock while we're still servicing an interrupt.
> 
> Actually, this explanation is not correct - it should be more like:
> 
> serial8250_startup() doesn't disable interrupts while taking the 
> &up->port.lock which might race against the interrupt handler 
> serial8250_interrupt(), which when entered, will deadlock waiting for 
> the lock to be released.

thanks Borislav, your patch seems to have done the trick.

Tested-by: Ingo Molnar <mingo@elte.hu>

	Ingo