[GIT PULL v2 rcu/next] memory-barriers.txt commits for 4.6

[GIT PULL v2 rcu/next] memory-barriers.txt commits for 4.6

[Paul E. McKenney]

Hello, Ingo,

This series consists of changes to memory-barriers.txt, that is, the
contents of the earlier documentation-series pull request, but leaving
out the controversial Documentation/RCU/Design changes.

I do not expect to have Documentation/RCU/Design commits for the
current merge window.  In fact, in view of the no-diagrams and
no-quizzes restrictions, I don't see a way to improve on comments
in the source code.  I therefore I expect that I will drop that work
entirely.  If anyone has any suggestions on what to do with the existing
Documentation/RCU/Design/Requirements, up to and including "git rm",
please let me know.

The memory-barriers.txt commits are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git for-mingo

for you to fetch changes up to 65f95ff2e41a32dd190cf28e3abb029625eef968:

  documentation: Clarify compiler store-fusion example (2016-03-14 15:52:19 -0700)

----------------------------------------------------------------
Paul E. McKenney (7):
      documentation: Fix control dependency and identical stores
      documentation: Fix memory-barriers.txt section references
      documentation: Remove obsolete reference to RCU-protected indexes
      documentation: Subsequent writes ordered by rcu_dereference()
      documentation: Distinguish between local and global transitivity
      documentation:  Add alternative release-acquire outcome
      documentation: Transitivity is not cumulativity

SeongJae Park (1):
      documentation: Clarify compiler store-fusion example

 Documentation/memory-barriers.txt | 141 +++++++++++++++++++++++++++++++-------
 1 file changed, 116 insertions(+), 25 deletions(-)

Re: [GIT PULL v2 rcu/next] memory-barriers.txt commits for 4.6

[Ingo Molnar]

* Paul E. McKenney <paulmck@linux.vnet.ibm.com> wrote:

> [...] In fact, in view of the no-diagrams and no-quizzes restrictions, I don't 
> see a way to improve on comments in the source code. [...]

So I don't think there's such a hard restriction: AFAICS Linus's main problem was 
the dual .html and .htmlx file for what essentially are the same documents.

If you can solve that with a single (.html?) file then I think it's all OK.

> The memory-barriers.txt commits are available in the git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git for-mingo
> 
> for you to fetch changes up to 65f95ff2e41a32dd190cf28e3abb029625eef968:
> 
>   documentation: Clarify compiler store-fusion example (2016-03-14 15:52:19 -0700)
> 
> ----------------------------------------------------------------
> Paul E. McKenney (7):
>       documentation: Fix control dependency and identical stores
>       documentation: Fix memory-barriers.txt section references
>       documentation: Remove obsolete reference to RCU-protected indexes
>       documentation: Subsequent writes ordered by rcu_dereference()
>       documentation: Distinguish between local and global transitivity
>       documentation:  Add alternative release-acquire outcome
>       documentation: Transitivity is not cumulativity
> 
> SeongJae Park (1):
>       documentation: Clarify compiler store-fusion example
> 
>  Documentation/memory-barriers.txt | 141 +++++++++++++++++++++++++++++++-------
>  1 file changed, 116 insertions(+), 25 deletions(-)

Pulled into tip:locking/urgent, thanks a lot Paul!

	Ingo

Re: [GIT PULL v2 rcu/next] memory-barriers.txt commits for 4.6

[Paul E. McKenney]

On Wed, Mar 16, 2016 at 08:26:25AM +0100, Ingo Molnar wrote:
> 
> * Paul E. McKenney <paulmck@linux.vnet.ibm.com> wrote:
> 
> > [...] In fact, in view of the no-diagrams and no-quizzes restrictions, I don't 
> > see a way to improve on comments in the source code. [...]
> 
> So I don't think there's such a hard restriction: AFAICS Linus's main problem was 
> the dual .html and .htmlx file for what essentially are the same documents.
> 
> If you can solve that with a single (.html?) file then I think it's all OK.

Very good!

I do have a prototype in -rcu on branch rcu/dev that uses Linus's second
suggestion, namely blanking out the answer so that the user mouse-selects
it to see it.  I am now chasing down the w3c validator complaints.
Of course, there will likely be some review feedback.  ;-)

> > The memory-barriers.txt commits are available in the git repository at:
> > 
> >   git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git for-mingo
> > 
> > for you to fetch changes up to 65f95ff2e41a32dd190cf28e3abb029625eef968:
> > 
> >   documentation: Clarify compiler store-fusion example (2016-03-14 15:52:19 -0700)
> > 
> > ----------------------------------------------------------------
> > Paul E. McKenney (7):
> >       documentation: Fix control dependency and identical stores
> >       documentation: Fix memory-barriers.txt section references
> >       documentation: Remove obsolete reference to RCU-protected indexes
> >       documentation: Subsequent writes ordered by rcu_dereference()
> >       documentation: Distinguish between local and global transitivity
> >       documentation:  Add alternative release-acquire outcome
> >       documentation: Transitivity is not cumulativity
> > 
> > SeongJae Park (1):
> >       documentation: Clarify compiler store-fusion example
> > 
> >  Documentation/memory-barriers.txt | 141 +++++++++++++++++++++++++++++++-------
> >  1 file changed, 116 insertions(+), 25 deletions(-)
> 
> Pulled into tip:locking/urgent, thanks a lot Paul!

Thank you, Ingo!

I will likely have one more locktorture RFC pull request, if that works
for you.  This fixes a bug and has very few users, so very low risk.

							Thanx, Paul