Re: linux-next: build warning after merge of the fbdev tree

[Laurent Pinchart <laurent.pinchart@ideasonboard.com>]

Hi Geert,

On Tuesday 20 December 2011 16:24:30 Geert Uytterhoeven wrote:
> On Tue, Dec 20, 2011 at 16:19, Laurent Pinchart wrote:
> > On Tuesday 20 December 2011 14:48:33 Geert Uytterhoeven wrote:
> >> On Tue, Dec 20, 2011 at 11:37, Laurent Pinchart wrote:
> >> > On Tuesday 20 December 2011 06:32:14 Stephen Rothwell wrote:
> >> >> After merging the fbdev tree, today's linux-next build (powerpc
> >> >> ppc64_defconfig) produced this warning:
> >> >> 
> >> >> drivers/video/matrox/matroxfb_base.c:150:2: warning: braces around
> >> >> scalar initializer [enabled by default]
> >> >> drivers/video/matrox/matroxfb_base.c:150:2: warning: (near
> >> >> initialization for 'vesafb_defined.colorspace') [enabled by default]
> >> >> drivers/video/matrox/matroxfb_base.c:150:2: warning: excess elements
> >> >> in scalar initializer [enabled by default]
> >> >> drivers/video/matrox/matroxfb_base.c:150:2: warning: (near
> >> >> initialization for 'vesafb_defined.colorspace') [enabled by default]
> >> >> drivers/video/matrox/matroxfb_crtc2.c:596:3: warning: braces around
> >> >> scalar initializer [enabled by default]
> >> >> drivers/video/matrox/matroxfb_crtc2.c:596:3: warning: (near
> >> >> initialization for 'matroxfb_dh_defined.colorspace') [enabled by
> >> >> default]
> >> >> drivers/video/matrox/matroxfb_crtc2.c:596:3: warning: excess elements
> >> >> in scalar initializer [enabled by default]
> >> >> drivers/video/matrox/matroxfb_crtc2.c:596:3: warning: (near
> >> >> initialization for 'matroxfb_dh_defined.colorspace') [enabled by
> >> >> default]
> >> >> 
> >> >> Introduced by commit fb21c2f42879 ("fbdev: Add FOURCC-based format
> >> >> configuration API").
> >> > 
> >> > The following patch should fix the issue. Florian, are you fine with
> >> > it, or would you rather modify the initializers to be name-based ?
> >> > 
> >> > From 078987fc14dba806fc1bc628e3e5c371db3cf1b8 Mon Sep 17 00:00:00 2001
> >> > From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> >> > Date: Tue, 20 Dec 2011 11:30:53 +0100
> >> > Subject: [PATCH] fbdev: matroxfb: Fix compilation after
> >> > fb_var_screeninfo change
> >> > 
> >> > Commit fb21c2f42879 ("fbdev: Add FOURCC-based format configuration
> >> > API") modified the layout of the fb_var_screeninfo structure. Update
> >> > the static initializers in the matroxfb driver accordingly.
> >> > 
> >> > Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> >> > ---
> >> >  drivers/video/matrox/matroxfb_base.c  |    2 +-
> >> >  drivers/video/matrox/matroxfb_crtc2.c |    2 +-
> >> >  2 files changed, 2 insertions(+), 2 deletions(-)
> >> > 
> >> > diff --git a/drivers/video/matrox/matroxfb_base.c
> >> > b/drivers/video/matrox/matroxfb_base.c index 44bf8d4..f98aad5 100644
> >> > --- a/drivers/video/matrox/matroxfb_base.c
> >> > +++ b/drivers/video/matrox/matroxfb_base.c
> >> > @@ -147,7 +147,7 @@ static struct fb_var_screeninfo vesafb_defined = {
> >> >        39721L,48L,16L,33L,10L,
> >> >        96L,2L,~0,      /* No sync info */
> >> >        FB_VMODE_NONINTERLACED,
> >> > -       0, {0,0,0,0,0}
> >> > +       0, 0, {0,0,0,0}
> >> 
> >> All these zeroes should not be specified, the compiler will do the
> >> right thing for
> >> static variables anyway.
> >> 
> >> Else they have to be updated manually everytime one of the reserved
> >> fields is consumed for a new feature.
> > 
> > I agree. Should I just remove them, or switch the structures to named
> > initializers while I'm at it ?
> 
> I would just remove them. What do they buy you?
> 
> Even with named initializers
> 
>     .reserved = { 0, 0, 0, 0 }
> 
> will break once we consume one more reserved field, unless you write
> 
>     .reserved = { 0, }
> 
> to let the compiler handle it, but then you could just not specify the
> field at all.

Sure. In both cases I would remove them, but I was wondering if I should 
switch the whole structure to named initializers in addition to removing them. 
I don't know if the common practice to solve issues in the -next tree is to 
apply patches that are as little intrusive as possible, or if bigger patches 
are allowed.

-- 
Regards,

Laurent Pinchart