From mboxrd@z Thu Jan  1 00:00:00 1970
From: Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH] kernel.h: Add for_each_if()
Date: Mon, 9 Jul 2018 16:30:01 -0700
Message-ID: <20180709163001.8fb8148223a57bc46a13fbda@linux-foundation.org>
References: <20180709083650.23549-1-daniel.vetter@ffwll.ch>
        <20180709162509.29343-1-daniel.vetter@ffwll.ch>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Return-path: <linux-kernel-owner@vger.kernel.org>
In-Reply-To: <20180709162509.29343-1-daniel.vetter@ffwll.ch>
Sender: linux-kernel-owner@vger.kernel.org
To: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: LKML <linux-kernel@vger.kernel.org>, DRI Development <dri-devel@lists.freedesktop.org>, Intel Graphics Development <intel-gfx@lists.freedesktop.org>, Gustavo Padovan <gustavo@padovan.org>, Maarten Lankhorst <maarten.lankhorst@linux.intel.com>, Sean Paul <seanpaul@chromium.org>, David Airlie <airlied@linux.ie>, Kees Cook <keescook@chromium.org>, Ingo Molnar <mingo@kernel.org>, Greg Kroah-Hartman <gregkh@linuxfoundation.org>, NeilBrown <neilb@suse.com>, Wei Wang <wvw@google.com>, Stefan Agner <stefan@agner.ch>, Andrei Vagin <avagin@openvz.org>, Randy Dunlap <rdunlap@infradead.org>, Andy Shevchenko <andriy.shevchenko@linux.intel.com>, Yisheng Xie <ysxie@foxmail.com>, Peter Zijlstra <peterz@infradead.org>, Daniel Vetter <daniel.vetter@intel.com>
List-Id: dri-devel@lists.freedesktop.org

On Mon,  9 Jul 2018 18:25:09 +0200 Daniel Vetter <daniel.vetter@ffwll.ch> wrote:

> To avoid compilers complainig about ambigious else blocks when putting
> an if condition into a for_each macro one needs to invert the
> condition and add a dummy else. We have a nice little convenience
> macro for that in drm headers, let's move it out. Subsequent patches
> will roll it out to other places.
> 
> The issue the compilers complain about are nested if with an else
> block and no {} to disambiguate which if the else belongs to. The C
> standard is clear, but in practice people forget:
> 
> if (foo)
> 	if (bar)
> 		/* something */
> 	else
> 		/* something else

um, yeah, don't do that.  Kernel coding style is very much to do

	if (foo) {
		if (bar)
			/* something */
		else
			/* something else
	}

And if not doing that generates a warning then, well, do that.

> The same can happen in a for_each macro when it also contains an if
> condition at the end, except the compiler message is now really
> confusing since there's only 1 if:
> 
> for_each_something()
> 	if (bar)
> 		/* something */
> 	else
> 		/* something else
> 
> The for_each_if() macro, by inverting the condition and adding an
> else, avoids the compiler warning.

Ditto.

> Motivated by a discussion with Andy and Yisheng, who want to add
> another for_each_macro which would benefit from for_each_if() instead
> of hand-rolling it.

Ditto.

> v2: Explain a bit better what this is good for, after the discussion
> with Peter Z.

Presumably the above was discussed in whatever-thread-that-was.