Re: [PATCH 0/3 v3] rcu: Detect rcu uses under extended quiescent state

On Fri, Jun 24, 2011 at 11:18:45AM +0200, Peter Zijlstra wrote:
> On Fri, 2011-06-24 at 01:12 +0200, Frederic Weisbecker wrote:
> > This time I have no current practical cases to fix. Those I fixed
> > in previous versions were actually using rcu_dereference_raw(), which
> > is legal in extended qs.
> > 
> > Frederic Weisbecker (3):
> >   rcu: Detect illegal rcu dereference in extended quiescent state
> >   rcu: Inform the user about dynticks idle mode on PROVE_RCU warning
> >   rcu: Warn when rcu_read_lock() is used in extended quiescent state
> > 
> >  include/linux/rcupdate.h |   68 +++++++++++++++++++++++++++++++++++++++-------
> >  kernel/lockdep.c         |    4 +++
> >  kernel/rcupdate.c        |    4 +++
> >  kernel/rcutiny.c         |   13 +++++++++
> >  kernel/rcutree.c         |   14 +++++++++
> >  5 files changed, 93 insertions(+), 10 deletions(-)
> 
> Right, so the only comment I have is that it might have been nice to
> explain what the heck an extended qs is :-) I've since figured it out,
> but it did require waking up my brain.

Hehe, right :)

As I have scratched my head for some times with rcu dynticks for the nohz cpuset
things, I guess that in the meantime my unconscious mind developed the idea that rcu extended
quiescent states were a worldwide topic that every people talk about in dinner with their family,
that these even became the core stories of some lullabies. I can still hear its chorus, that makes
one switching to idle peacefully...