On Fri, 16 Jul 2010, luca ellero wrote:
> Robert P. J. Day ha scritto:
> > one of my current students observes the following -- that the proc
> > files /proc/irq/<PID>/smp_affinity are not only readable, but writable
> > and wonders why one would want to *change* the smp_affinity of IRQs
> > for a given process (if that's actually what that would mean).
> >
> > i've never looked closely at that. any thoughts?
> >
> > rday
> >
> >
> ASAIK PID here is not "Process ID" but is the "irq ID". So
> smp_affinity simply binds that irq to a particular CPU.
you're right, i typed that when not fully awake yet, sorry.
> This is done for better cache locality I think, because this irq
> will be always served by the same CPU and the handler will
> (hopefully) be in the local cache. Please correct me if I'm wrong.
> regards
thanks, i think i see how it works now. will investigate further
later.
rday
--
========================================================================
Robert P. J. Day Waterloo, Ontario, CANADA
Top-notch, inexpensive online Linux/OSS/kernel courses
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to [email protected]
Please read the FAQ at http://kernelnewbies.org/FAQ
