On Wed, Oct 05, 2005 at 09:14:58PM +1000, Paul Mackerras was heard to remark:
> Linas writes:
>
> > 03-eeh-statistics.patch
>
> > + if (!dn) {
> > + __get_cpu_var(no_dn)++;
>
> We have to make sure we are not preemptible when we use
> __get_cpu_var, since it uses smp_processor_id(). It's not clear to me
> that we have ensured that in every case where we use __get_cpu_var.
> Are you sure that we hold a spinlock, or are at interrupt level, or
> have explicitly disabled preemption at every point where we use
> __get_cpu_var?
Tese used to be plain-old global variables, but someone submitted
a patch that to turn them into the __get_cpu_var() form. I don't
know why; there's no real performance reason, since these are almost
never incremented, except a bit during boot. What if we just change
them back to global vars?
I've also day-dreamed about moving these stats to somewhere in
in the /sys directory. Any suggestions there?
--linas
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]
