[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <045de91c8b8e6c8ca60b2c4baf4430bd5cd5eeea.camel@perches.com>
Date: Thu, 16 May 2024 17:37:39 -0700
From: Joe Perches <joe@...ches.com>
To: jim.cromie@...il.com
Cc: linux-kernel@...r.kernel.org, akpm@...uxfoundation.org, Andy Whitcroft
<apw@...onical.com>, Dwaipayan Ray <dwaipayanray1@...il.com>, Lukas Bulwahn
<lukas.bulwahn@...il.com>
Subject: Re: [PATCH 1/1] checkpatch: allow multi-statement declarative
macros.
On Thu, 2024-05-16 at 09:20 -0600, jim.cromie@...il.com wrote:
> On Thu, May 16, 2024 at 8:43 AM Joe Perches <joe@...ches.com> wrote:
> >
> > On Thu, 2024-05-16 at 08:14 -0600, Jim Cromie wrote:
> > > Declarative macros, which declare/define storage (at either file or
> > > function scope), cannot be wrapped in do-while statements. So
> > > checkpatch advice is incorrect here.
> > >
> > > The code has an $exceptions regex which allows multiple statements
> > > based on the macro name, etc; /DECLARE_PER_CPU|DEFINE_PER_CPU/ are
> > > currently accepted, widen those to accept /DECLARE|DEFINE/.
> >
> > It seems this exempts too large a number of these macros
> >
> > $ git grep -P '^\s*\#\s*define\s+\w*(?:DECLARE|DEFINE)\w*'|wc -l
> > 5075
> >
>
> wow, thats more than Id have thought.
>
> > How about somehow limiting these exemptions more strictly?
>
> agreed. I'll just add my 1 exceptional macro name.
> resending shortly.
Is this macro used in a lot of places?
Otherwise, why not just ignore the macro where it occurs?
Powered by blists - more mailing lists