Re: [watchdog] combine nmi_watchdog and softlockup

[Don Zickus <dzickus@redhat.com>]

On Tue, Apr 06, 2010 at 12:11:11AM +0400, Cyrill Gorcunov wrote:
> On Tue, Mar 30, 2010 at 10:52:38AM -0400, Aristeu Sergio Rozanski Filho wrote:
> > > On Sat, Mar 27, 2010 at 10:46:50PM -0400, Aristeu Sergio Rozanski Filho wrote:
> > > > Hi Don,
> > > > > +/* deprecated */
> > > > > +static int __init nosoftlockup_setup(char *str)
> > > > > +{
> > > > > +	no_watchdog = 1;
> > > > > +	return 1;
> > > > > +}
> > > > > +__setup("nosoftlockup", nosoftlockup_setup);
> > > > > +static int __init nonmi_watchdog_setup(char *str)
> > > > > +{
> > > > > +	no_watchdog = 1;
> > > > > +	return 1;
> > > > > +}
> > > > > +__setup("nonmi_watchdog", nonmi_watchdog_setup);
> > > > didn't you just add nonmi_watchdog parameter? I don't think there's a reason
> > > > to keep compatibility here.
> > > 
> > > Hmm, I think you are right.  I thought I added that because it existed in
> > > the old nmi_watchdog setup but I can't find it.  So yeah, I can drop that.
> > you could provide a nmi_watchdog=0 backwards compatibility and warn about
> > values != 0
> > 
> > -- 
> > Aristeu
> > 
> 
> Sorry for a long delay, I think we might need to inform a user that "lapic",
> "ioapic" is no longer used (perf-nmi is supposed to substitute the former nmi
> code in a long term right?) so that for some time period, say the whole release
> cycle, if lapic or ioapic, or numbers are passed to nmi_watchdog= setup option
> we would just print out that the parameters are deprecated and better to not
> use them any longer. Hm?

Agreed.

Cheers,
Don