Re: [PATCH 4/6] staging: sm750fb: correct incompatible pointer type

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Tue, Mar 10, 2015 at 10:46:55PM +0530, Sudip Mukherjee wrote:
> we were getting build warnings about assignment of incompatible
> pointer types. some of the function definitions were having wrong
> return type or arguments.
> 
> Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> ---
>  drivers/staging/sm750fb/sm750.h       | 11 ++++++-----
>  drivers/staging/sm750fb/sm750_accel.c |  4 ++--
>  drivers/staging/sm750fb/sm750_accel.h |  4 ++--
>  3 files changed, 10 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/staging/sm750fb/sm750.h b/drivers/staging/sm750fb/sm750.h
> index d39968c..edb955a 100644
> --- a/drivers/staging/sm750fb/sm750.h
> +++ b/drivers/staging/sm750fb/sm750.h
> @@ -23,7 +23,7 @@ struct lynx_accel{
>  	volatile unsigned char __iomem * dpPortBase;
>  
>  	/* function fointers */
> -	int (*de_init)(struct lynx_accel *);
> +	void (*de_init)(struct lynx_accel *);

That's ok, but:

>  
>  	int (*de_wait)(void);/* see if hardware ready to work */
>  
> @@ -34,8 +34,8 @@ struct lynx_accel{
>  						u32,u32,u32,u32,
>  						u32,u32,u32,u32);
>  
> -	int (*de_imageblit)(struct lynx_accel *,const char *,u32,u32,u32,
> -						u32,u32,u32,u32,u32,u32,u32,u32,u32);
> +	int (*de_imageblit)(struct lynx_accel *, const char *, u32, u32, u32,
> +			    u32, u32, u32, u32, u32, u32, u32, u32, u32);
>  


This isn't fixing an error.

>  };
>  
> @@ -120,8 +120,9 @@ struct lynxfb_crtc{
>  	int(*proc_setColReg)(struct lynxfb_crtc*,ushort,ushort,ushort,ushort);
>  	void (*clear)(struct lynxfb_crtc*);
>          /* pan display */
> -        int(*proc_panDisplay)(struct lynxfb_crtc*, struct fb_var_screeninfo*,
> -                struct fb_info*);
> +	int (*proc_panDisplay)(struct lynxfb_crtc*,
> +			       const struct fb_var_screeninfo*,
> +			       const struct fb_info*);

That's a nice cleanup, but it's not even the correct cleanup.

Please just fix the specific warning, don't mix what you are doing in
one patch.  You aren't documenting this in the changelog information, so
I have to reject it, sorry.

greg k-h