lkml.org 
[lkml]   [2012]   [Jun]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 1/2] x86: Track minimum microcode revision globally v2
On Wed, Jun 13, 2012 at 10:49:52AM +0200, Peter Zijlstra wrote:
> On Tue, 2012-06-12 at 14:25 -0700, Andi Kleen wrote:
> > +void update_min_microcode(struct cpuinfo_x86 *c)
> > +{
> > + int i;
> > +
> > + for_each_online_cpu (i)
>
> Superfluous whitespace
>
> > + if (cpu_data(i).microcode < c->microcode)
> > + return;
>
> That needs {}

You must be following a different code style guide than the Linux one.


> > c->microcode = val[1];
> >
> > + update_min_microcode(c);
> > return 0;
> > }
>
> Doing it here means doing the for_each_cpu thing with preempt/irqs
> disabled, that's not funny.

Well a ucode update is a really slow operation anyways. And the loop
gets stopped at the first mismatch. So it'll never be n^2 and in most
cases much faster. Generally the loop should be several orders
of magnitude less than the actual cost of the update, even on large
systems.

> Also this is still a O(n^2) proposition.. so how is this better than the
> notifier thing I had?

Simpler at least.

I don't know why people love notifiers, they are a "COME FROM" and make
every code who uses them a mess.

As for CPU_STARTING don't know how complicated it would be. I suppose
it could be done as a follow up.

-Andi

--
ak@linux.intel.com -- Speaking for myself only


\
 
 \ /
  Last update: 2012-06-13 17:21    [from the cache]
©2003-2014 Jasper Spaans. Advertise on this site