Re: [PATCH v4 2/2] LAM: Enable tagged pointer support for watchpoints.

[Eli Zaretskii <eliz@gnu.org>]

> From: "Schimpe, Christina" <christina.schimpe@intel.com>
> CC: "gdb-patches@sourceware.org" <gdb-patches@sourceware.org>, "Willgerodt,
>  Felix" <felix.willgerodt@intel.com>, "luis.machado@arm.com"
> 	<luis.machado@arm.com>
> Date: Thu, 4 Jul 2024 13:05:35 +0000
> 
> > > diff --git a/gdb/NEWS b/gdb/NEWS
> > > index 47677cb773a..fcbb89e8bf8 100644
> > > --- a/gdb/NEWS
> > > +++ b/gdb/NEWS
> > > @@ -3,6 +3,8 @@
> > >
> > >  *** Changes since GDB 15
> > >
> > > +* GDB now supports watchpoints for tagged data pointers on amd64.
> > > +
> > >  * Debugger Adapter Protocol changes
> > 
> > Thanks.  The above NEWS text is okay, but I wonder whether we explain
> > anywhere what are "tagged data pointers".
> > 
> > Reviewed-By: Eli Zaretskii <eliz@gnu.org>
> 
> Thanks for the review. I understand this can be a bit confusing. 
> Is " GDB now supports watchpoints for tagged pointers used for data accesses on amd64." better?
> Or, we could also just write: " GDB now supports watchpoints for tagged pointers on amd64.", as watchpoints are usually just data breakpoints.
> 
> What do you think?

My bother is actually about what we say in the manual, not in NEWS.
If the manual explains this terminology, it is fine to leave it
unexplained in NEWS, as users are expected to turn to the manual for
the details of the NEWS entries they are interested in.  But if this
is not explained in the manual, then we should either explain it, or
do that in NEWS entry itself, and in the latter case I'm afraid your
suggestion above is not enough because "tagged pointers" are not
explained anywhere, either.