[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1551974540.9796.69.camel@acm.org>
Date: Thu, 07 Mar 2019 08:02:20 -0800
From: Bart Van Assche <bvanassche@....org>
To: Arnd Bergmann <arnd@...db.de>,
Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...hat.com>,
Will Deacon <will.deacon@....com>
Cc: Waiman Long <longman@...hat.com>,
"Steven Rostedt (VMware)" <rostedt@...dmis.org>,
Tetsuo Handa <penguin-kernel@...ove.SAKURA.ne.jp>,
Frederic Weisbecker <frederic@...nel.org>,
"Joel Fernandes (Google)" <joel@...lfernandes.org>,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] lockdep: avoid a clang warning
On Thu, 2019-03-07 at 08:52 +0100, Arnd Bergmann wrote:
> Clang warns about a tentative array definition without a length:
>
> kernel/locking/lockdep.c:845:12: error: tentative array definition assumed to have one element [-Werror]
>
> There is no real reason to do this here, so just set the same length as
> in the real definition later in the same file. It has to be hidden in
> an #ifdef or annotated __maybe_unused though, to avoid the unused-variable
> warning if CONFIG_PROVE_LOCKING is disabled.
>
> Signed-off-by: Arnd Bergmann <arnd@...db.de>
> ---
> kernel/locking/lockdep.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
> index 21cb81fe6359..35a144dfddf5 100644
> --- a/kernel/locking/lockdep.c
> +++ b/kernel/locking/lockdep.c
> @@ -842,7 +842,9 @@ static bool class_lock_list_valid(struct lock_class *c, struct list_head *h)
> return true;
> }
>
> -static u16 chain_hlocks[];
> +#ifdef CONFIG_PROVE_LOCKING
> +static u16 chain_hlocks[MAX_LOCKDEP_CHAIN_HLOCKS];
> +#endif
>
> static bool check_lock_chain_key(struct lock_chain *chain)
> {
Reviewed-by: Bart Van Assche <bvanassche@....org>
Powered by blists - more mailing lists