Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Mon, 29 Oct 2018 07:01:53 +0100
From:   Ingo Molnar <mingo@kernel.org>
To:     Carmeli Tamir <carmeli.tamir@gmail.com>
Cc:     mingo@redhat.com, peterz@infradead.org,
        linux-kernel@vger.kernel.org, Thomas Gleixner <tglx@linutronix.de>,
        "Rafael J. Wysocki" <rjw@sisk.pl>, Tejun Heo <tj@kernel.org>
Subject: Re: [PATCH] scheduler: wait: Added missing include freezer.h
Message-ID: <20181029060153.GA128403@gmail.com>
References: <1540582551-44028-1-git-send-email-carmeli.tamir@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1540582551-44028-1-git-send-email-carmeli.tamir@gmail.com>
User-Agent: Mutt/1.9.4 (2018-02-28)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


* Carmeli Tamir <carmeli.tamir@gmail.com> wrote:

> The function 'wait_event_freezable' calls '__wait_event_freezable' that
> uses 'try_to_freeze', which is defined in freezer.h.
> This causes a compilation error for callers of 'wait_event_freezables',
> forcing them to include also freezer.h.
> 
> Signed-off-by: Carmeli Tamir <carmeli.tamir@gmail.com>
> ---
>  include/linux/wait.h | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/include/linux/wait.h b/include/linux/wait.h
> index ed7c122..75061ee 100644
> --- a/include/linux/wait.h
> +++ b/include/linux/wait.h
> @@ -4,6 +4,7 @@
>  /*
>   * Linux wait queue related types and methods
>   */
> +#include <linux/freezer.h>
>  #include <linux/list.h>
>  #include <linux/stddef.h>
>  #include <linux/spinlock.h>

I don't think we should do this, because freezer.h then includes 
<linux/sched.h> - while much of the point of <linux/wait.h> was to not 
include the huge chain of dependencies that sched.h invokes...

BTW., another way this could be solved is to uninline try_to_freeze(), it 
seems a bit large at first sight to be inlined - but I could be wrong.

Once it's not inline we could just add this to wait.h:

  extern bool try_to_freeze(void);

or so. This decouples wait.h from freezer.h.

Thanks,

	Ingo