Re: Interactivity regression since v3.11 in mm/vmscan.c

[Michal Hocko <mhocko@suse.cz>]

On Fri 06-06-14 18:11:14, Felipe Contreras wrote:
> On Fri, Jun 6, 2014 at 5:33 AM, Felipe Contreras
> <felipe.contreras@gmail.com> wrote:
> > On Fri, Jun 6, 2014 at 4:16 AM, Michal Hocko <mhocko@suse.cz> wrote:
> >
> >> Mel has a nice systemtap script (attached) to watch for stalls. Maybe
> >> you can give it a try?
> >
> > Is there any special configurations I should enable?
> >
> > I get this:
> > semantic error: unresolved arity-1 global array name, missing global
> > declaration?: identifier 'name' at /tmp/stapd6pu9A:4:2
> >         source: name[t]=execname()
> >                 ^
> >
> > Pass 2: analysis failed.  [man error::pass2]
> > Number of similar error messages suppressed: 71.
> > Rerun with -v to see them.
> > Unexpected exit of STAP script at
> > /home/felipec/Downloads/watch-dstate-new.pl line 320.
> 
> Actually I debugged the problem, and it's that the format of the
> script is DOS, not UNIX. After changing the format the script works.

Ups, I've downloaded it from our bugzilla so maybe it just did some
tricks with the script.

> However, it's not returning anything. It's running, but doesn't seem
> to find any stalls.

Intereting. It was quite good at pointing at stalls. How are you
measuring those stalls during your testing?
-- 
Michal Hocko
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>