Messages in this thread | | | Subject | Re: [(RT RFC) PATCH v2 6/9] add a loop counter based timeout mechanism | From | Sven-Thorsten Dietrich <> | Date | Mon, 25 Feb 2008 14:39:29 -0800 |
| |
On Mon, 2008-02-25 at 23:06 +0100, Pavel Machek wrote: > On Mon 2008-02-25 11:01:08, Gregory Haskins wrote: > > From: Sven Dietrich <sdietrich@novell.com> > > Why is this good idea? >
The timeout is useful to eliminate excessive CPU utilization when waiting for long-held critical sections.
The patch header should state this :) Will fix.
> > Signed-off-by: Sven Dietrich <sdietrich@novell.com> > > --- > > > > > +config RTLOCK_DELAY > > + int "Default delay (in loops) for adaptive rtlocks" > > + range 0 1000000000 > > + depends on ADAPTIVE_RTLOCK > > + default "10000" > > + help > > + This allows you to specify the maximum attempts a task will spin > > + attempting to acquire an rtlock before sleeping. The value is > > + tunable at runtime via a sysctl. A setting of 0 (zero) disables > > + the adaptive algorithm entirely. > > + > > I believe you have _way_ too many config variables. If this can be set > at runtime, does it need a config option, too? >
Absolutely. The sysctl was added after-the-fact, so this is a relic, we will remove the config option, and define the default timeout in the appropriate header.
Sven
> Pavel
| |