[linux-dvb] [BUG] flexcop lockdep
xyzzy at speakeasy.org
Fri Apr 6 20:34:44 CEST 2007
On Fri, 6 Apr 2007, Patrick Boettcher wrote:
> On Thu, 5 Apr 2007, Trent Piepho wrote:
> > On Thu, 5 Apr 2007, Patrick Boettcher wrote:
> > > On Thu, 5 Apr 2007, Borgi2008 wrote:
> > > > Am Mittwoch, den 04.04.2007, 23:29 +0300 schrieb Antti Sepp=E4l=E4:
> > > > >
> > > > > Actually, looking at the code I cannot figure out why there has t=
o be a
> > > > > spinlock in the first place.
> > > > >
> > > > > The lock is only taken in the interrupt handler which already run=
> > > > > atomic context so there is no use in making the handler protected=
> > > > > spinlock. Am I missing something here?
> > > > >
> > > > > I think the spinlock is unnecessary and should be removed entirel=
> > >
> > > Even on SMP systems? ISRs are only atomic on one CPU.
> > I thought ISRs were normally atomic even on SMP systems. When an inter=
> > occurs, that interrupt is disabled until the ISR returns. Except in fa=
> > handlers (which flexcop is not) all interrupts aren't disabled, and so =
> > ISR can be interrupted by a _different_ ISR, and a different ISR could =
> > running at the same time on another CPU. But an ISR doesn't have to wo=
> > about two copies of itself running at the same time.
> > At least, that's how I think it works.
> > I don't see how a spin lock that is only used from the ISR could be use=
> > assuming the ISR is only called via an interrupt. There is no reason y=
> > couldn't call the isr function from some system call handler, but that
> > would be an unusual thing to do.
> > Normally an ISR does need a spinlock, to access data shared by the
> > non-interrupt part of the driver. I wonder if there is a bug in flexco=
> > that nothing else uses irq_lock?
> You can only take a spinlock at IRQ-level and not at normal kernel level.
Sure you can. There is a ton of code that does this. It would be pretty
much impossible to create SMP safe ISRs if there was no locking mechanism
that worked across interrupt-level to kernel-level.
Look in Documentation/cli-sti-removal.txt:
unsigned long flags;
driver_data.finish =3D 1;
driver_data.new_work =3D 0;
driver_data.finish =3D 0;
driver_data.new_work =3D 2;
More information about the linux-dvb