[PATCH] hvc_console polling mode timer backoff

Michael Ellerman michael at ellerman.id.au
Sun Apr 15 23:33:32 EST 2007


On Fri, 2007-04-13 at 14:11 -0500, Will Schmidt wrote:
> On Fri, 2007-13-04 at 17:47 +1000, Michael Ellerman wrote:
> > Because this is a super tight performance critical inner loop I've
> > changed it to use a shift instead of divide. So now it's more like 1.5%,
> > but still the point is it increases faster as time passes.
> 
> Using a shift here is a good idea.  I had considered accelerating the
> backoff value, but it got more complex than I wanted quickly.. I didnt
> consider a shift. :-) 
> 
> > It'd be good if some more people can bang on this and see what they
> > think of the interactivity.
> 
> Still works OK on my power4.   I can see the delay on my console, but
> I'm specifically looking for it, I don't think it's an issue. 

Cool. Yeah I think people will hardly notice it unless they're trying to
t notice it.

> > +#include <linux/debugfs.h>
> > +static int hvc_debug_init(void)
> > +{
> > +	debugfs_create_u32("hvc_timeout", 0400, NULL, &timeout);
> > +	return 0;
> > +}
> > +__initcall(hvc_debug_init);
> > +
> 
> Is this part temporary for your graphing, or think it's something that
> should go in? 

No that's just debug foobar I used for the graphs, I'll send a cleaned
up version during the week if you don't beat me to it.

> I used /proc/timer_stats output for my graphing, just dont have a good
> public spot to upload the pics.. 

Time to get a blag! ;)

cheers

-- 
Michael Ellerman
OzLabs, IBM Australia Development Lab

wwweb: http://michael.ellerman.id.au
phone: +61 2 6212 1183 (tie line 70 21183)

We do not inherit the earth from our ancestors,
we borrow it from our children. - S.M.A.R.T Person
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.ozlabs.org/pipermail/linuxppc-dev/attachments/20070415/237ecd04/attachment.pgp>


More information about the Linuxppc-dev mailing list