On Wed, Jul 21, 2021 at 02:36:30AM +0800, Zhouyi Zhou wrote:
> Thanks Paul for reviewing the patch
>
> On Wed, Jul 21, 2021 at 1:33 AM Paul E. McKenney <paulmck@xxxxxxxxxx> wrote:
> >
> > On Tue, Jul 20, 2021 at 03:17:25PM +0800, Zhouyi Zhou wrote:
> > > Hi, Paul
> > > I think preposition 'on' here is superfluous, correct me if I am wrong.
> > > Best Wishes
> > > Zhouyi
> >
> > The current text is technically grammatically correct, but as you
> > noticed, rather awkward. I should rework that sentence and the one
> > following. Huh. Actually, I should rework that entire paragraph.
> >
> > How about the following?
> >
> > Thanx, Paul
> >
> > ------------------------------------------------------------------------
> >
> > commit 5eb7525e2f7338ea6b817c145a9b4da60e6871d1
> > Author: Paul E. McKenney <paulmck@xxxxxxxxxx>
> > Date: Tue Jul 20 10:31:03 2021 -0700
> >
> > defer/rcufundamental: Wordsmith QQ answer about writes in RCU readers
> >
> > Reported-by: Zhouyi Zhou <zhouzhouyi@xxxxxxxxx>
> > Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
> >
> > diff --git a/defer/rcufundamental.tex b/defer/rcufundamental.tex
> > index 66f9b702..054512e0 100644
> > --- a/defer/rcufundamental.tex
> > +++ b/defer/rcufundamental.tex
> > @@ -330,12 +330,17 @@ greater detail in \cref{sec:memorder:RCU}.
> > \co{P1()}'s grace period.
> >
> > It might seem strange to have RCU read-side critical sections
> > - containing writes, but RCU is just fine with this.
> > - This capability is used frequently in the Linux kernel, for
> > - example, acquiring a lock on or reference to a data structure.
> > - Acquiring either a lock or a reference results in a write
> > - to memory, and it is OK to do these within an RCU read-side
> > - critical section.
> > + containing writes, but this capability is not only permitted,
> > + but also highly useful.
> > + For example, the Linux kernel frequently carries out an
> > + RCU-protected traversal of a linked data structure and then
> > + acquires a reference to the destination data element.
> > + Because this data element must not be freed in the meantime,
> > + that element's reference counter must necessarily be incremented
> > + within the traversal's RCU read-side critical section.
> > + However, that increment entails a write to memory.
> > + Therefore, it is a very good thing that memory writes are
> > + permitted within RCU read-side critical sections.
> The revised paragraph is easier for me to understand, especially why acquiring
> a reference to a data structure entails a write ;-)
> >
> > If having writes in RCU read-side critical sections still seems
> > strange, please review
> The revised edition looks wonderful, thank you
Glad you like it, and thank you for pointing me at the original
awkward sentence!
Thanx, Paul
