[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <YQ0abcsiQBr7y0sT@smile.fi.intel.com>
Date: Fri, 6 Aug 2021 14:18:05 +0300
From: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
To: John Ogness <john.ogness@...utronix.de>
Cc: Petr Mladek <pmladek@...e.com>,
Sergey Senozhatsky <senozhatsky@...omium.org>,
Steven Rostedt <rostedt@...dmis.org>,
Thomas Gleixner <tglx@...utronix.de>,
linux-kernel@...r.kernel.org,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Masahiro Yamada <masahiroy@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
Nick Desaulniers <ndesaulniers@...gle.com>,
"Paul E. McKenney" <paulmck@...nel.org>,
Tetsuo Handa <penguin-kernel@...ove.sakura.ne.jp>,
Vitor Massaru Iha <vitor@...saru.org>,
Sedat Dilek <sedat.dilek@...il.com>,
Changbin Du <changbin.du@...el.com>
Subject: Re: [PATCH printk v1 07/10] console: add write_atomic interface
On Fri, Aug 06, 2021 at 01:02:40PM +0206, John Ogness wrote:
> On 2021-08-03, Andy Shevchenko <andriy.shevchenko@...ux.intel.com> wrote:
...
> >> + if (!(con->flags & CON_ENABLED)) \
> >> + continue; \
> >
> > What about
> >
> > #define console_is_enabled(con) (!!(con->flags & CON_ENABLED))
> >
> > or inliner equivalent
> >
> > static inline bool console_is_enabled(struct console *con)
> > {
> > return !!(con->flags & CON_ENABLED);
> > }
>
> Generally kernel code uses the console flags directly. A quick check for
> CON_ENABLED shows direct flag queries all over:
>
> $ git grep -l -e 'flags.*& .*CON_ENABLED' | wc -c
> 16
>
> Are you suggesting I replace this usage in all of these files? Or just
> the one macro in console.h for now? And since there are 6 more console
> flags, they should probably also have equivalent wrappers?
Introduce it now and reuse in new code, but somebody may clean up the rest in
the future.
--
With Best Regards,
Andy Shevchenko
Powered by blists - more mailing lists