[PATCH] ARM: OMAP2+: Fix compile error when FB_OMAP2 is not set

Tony Lindgren tony at atomide.com
Fri Mar 9 13:26:10 EST 2012


* Tomi Valkeinen <tomi.valkeinen at ti.com> [120309 00:14]:
> On Thu, 2012-03-08 at 12:46 -0800, Tony Lindgren wrote:
> > * Tony Lindgren <tony at atomide.com> [120308 12:11]:
> > > Otherwise we will get:
> > > 
> > > arch/arm/plat-omap/fb.c:101: error: expected ‘)’ before ‘*’ token
> > > 
> > > Signed-off-by: Tony Lindgren <tony at atomide.com>
> > > 
> > > ---
> > > 
> > > Tomi, you need something like this for what you have in for-next.
> > > 
> > > --- a/arch/arm/plat-omap/fb.c
> > > +++ b/arch/arm/plat-omap/fb.c
> > > @@ -98,6 +98,8 @@ arch_initcall(omap_init_fb);
> > >  
> > >  #else
> > >  
> > > -void __init omapfb_set_lcd_config(omap_lcd_config *config) { }
> > > +void __init omapfb_set_lcd_config(const struct omap_lcd_config *config)
> > > +{
> > > +}
> > >  
> > >  #endif
> > 
> > Also now seeing the following with what you have in for-next:
> > 
> > drivers/video/omap2/displays/panel-taal.c: In function ‘taal_num_errors_show’:
> > drivers/video/omap2/displays/panel-taal.c:605: warning: ‘errors’ may be used uninitialized in this function
> 
> Hmm, my compiler doesn't complain (codesourcery 2010.09-50), so I didn't
> notice that. Which compiler version reports the warning?
> 
> I'm not 100% sure, but I think it's a false warning, the code path that
> uses "errors" should only be ran when "errors" has been set.

This seems to depend on the .config options somehow. Not getting
it with omap2plus_defconfig, but getting it at least with Russell's
omap4430-sdp seed config from:

http://www.arm.linux.org.uk/developer/build/

Using now emdebian 4.3.5-4 as there seems to be some issues
showing all the section warnings with CodeSourcery and Linaro
toolchains.

Regards,

Tony



More information about the linux-arm-kernel mailing list