Re: [PATCH] m32r: fix readl/writel prototypes

[Luc Van Oostenryck]

On Tue, Jan 02, 2018 at 11:56:41AM +0100, Arnd Bergmann wrote:
> All other architectures use 'unsigned int' as the data in readl/write,
> but m32r uses 'unsigned long', leading to lots of harmless build warnings
> like:
> 
> drivers/mmc/host/dw_mmc.c: In function 'dw_mci_regs_show':
> drivers/mmc/host/dw_mmc.c:168:31: warning: format '%x' expects argument of 
> type 'unsigned int', but argument 3 has type 'long unsigned int' [-Wformat=]
>   seq_printf(s, "STATUS:\t0x%08x\n", mci_readl(host, STATUS));
> 
> Signed-off-by: Arnd Bergmann 
> ---
>  arch/m32r/include/asm/io.h | 8 
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/m32r/include/asm/io.h b/arch/m32r/include/asm/io.h
> index 1b653bb16f9a..093486b5b931 100644
> --- a/arch/m32r/include/asm/io.h
> +++ b/arch/m32r/include/asm/io.h
> @@ -108,9 +108,9 @@ static inline unsigned short _readw(unsigned long addr)
>   return *(volatile unsigned short __force *)addr;
>  }
>  
> -static inline unsigned long _readl(unsigned long addr)
> +static inline unsigned int _readl(unsigned long addr)
>  {
> - return *(volatile unsigned long __force *)addr;
> + return *(volatile unsigned int __force *)addr;
>  }

The __force is not needed here, so I think it would be better to
remove it. Maybe better to do that in another patch, though.

-- Luc Van Oostenryck

Re: [PATCH] m32r: fix readl/writel prototypes

[Luc Van Oostenryck]

On Tue, Jan 02, 2018 at 11:56:41AM +0100, Arnd Bergmann wrote:
> All other architectures use 'unsigned int' as the data in readl/write,
> but m32r uses 'unsigned long', leading to lots of harmless build warnings
> like:
> 
> drivers/mmc/host/dw_mmc.c: In function 'dw_mci_regs_show':
> drivers/mmc/host/dw_mmc.c:168:31: warning: format '%x' expects argument of 
> type 'unsigned int', but argument 3 has type 'long unsigned int' [-Wformat=]
>   seq_printf(s, "STATUS:\t0x%08x\n", mci_readl(host, STATUS));
> 
> Signed-off-by: Arnd Bergmann 
> ---
>  arch/m32r/include/asm/io.h | 8 
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/m32r/include/asm/io.h b/arch/m32r/include/asm/io.h
> index 1b653bb16f9a..093486b5b931 100644
> --- a/arch/m32r/include/asm/io.h
> +++ b/arch/m32r/include/asm/io.h
> @@ -108,9 +108,9 @@ static inline unsigned short _readw(unsigned long addr)
>   return *(volatile unsigned short __force *)addr;
>  }
>  
> -static inline unsigned long _readl(unsigned long addr)
> +static inline unsigned int _readl(unsigned long addr)
>  {
> - return *(volatile unsigned long __force *)addr;
> + return *(volatile unsigned int __force *)addr;
>  }

The __force is not needed here, so I think it would be better to
remove it. Maybe better to do that in another patch, though.

-- Luc Van Oostenryck

Re: [PATCH] m32r: fix readl/writel prototypes

[Geert Uytterhoeven]

Hi Arnd,

On Tue, Jan 2, 2018 at 11:56 AM, Arnd Bergmann  wrote:
> All other architectures use 'unsigned int' as the data in readl/write,
> but m32r uses 'unsigned long', leading to lots of harmless build warnings
> like:
>
> drivers/mmc/host/dw_mmc.c: In function 'dw_mci_regs_show':
> drivers/mmc/host/dw_mmc.c:168:31: warning: format '%x' expects argument of 
> type 'unsigned int', but argument 3 has type 'long unsigned int' [-Wformat=]
>   seq_printf(s, "STATUS:\t0x%08x\n", mci_readl(host, STATUS));

Thanks for fixing this!

BTW, I started fixing this a while ago, too, but of course I dived too deep
into arch/m32r/platforms/*/io.c (there are plenty of more type inconsistencies
with asm-generic/io.h!), and gave up after ca. 500 lines of changes ;-)

> Signed-off-by: Arnd Bergmann 

With the below fixed:

Reviewed-by: Geert Uytterhoeven 

> --- a/arch/m32r/include/asm/io.h
> +++ b/arch/m32r/include/asm/io.h
> @@ -108,9 +108,9 @@ static inline unsigned short _readw(unsigned long addr)
> return *(volatile unsigned short __force *)addr;
>  }
>
> -static inline unsigned long _readl(unsigned long addr)
> +static inline unsigned int _readl(unsigned long addr)
>  {
> -   return *(volatile unsigned long __force *)addr;
> +   return *(volatile unsigned int __force *)addr;
>  }
>
>  static inline void _writeb(unsigned char b, unsigned long addr)
> @@ -123,9 +123,9 @@ static inline void _writew(unsigned short w, unsigned 
> long addr)
> *(volatile unsigned short __force *)addr = w;
>  }
>
> -static inline void _writel(unsigned long l, unsigned long addr)
> +static inline void _writel(unsigned long l, unsigned int addr)

Don't you want to change the type of "l" instead of the type of "addr"?

>  {
> -   *(volatile unsigned long __force *)addr = l;
> +   *(volatile unsigned int __force *)addr = l;
>  }

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

Re: [PATCH] m32r: fix readl/writel prototypes

[Geert Uytterhoeven]

Hi Arnd,

On Tue, Jan 2, 2018 at 11:56 AM, Arnd Bergmann  wrote:
> All other architectures use 'unsigned int' as the data in readl/write,
> but m32r uses 'unsigned long', leading to lots of harmless build warnings
> like:
>
> drivers/mmc/host/dw_mmc.c: In function 'dw_mci_regs_show':
> drivers/mmc/host/dw_mmc.c:168:31: warning: format '%x' expects argument of 
> type 'unsigned int', but argument 3 has type 'long unsigned int' [-Wformat=]
>   seq_printf(s, "STATUS:\t0x%08x\n", mci_readl(host, STATUS));

Thanks for fixing this!

BTW, I started fixing this a while ago, too, but of course I dived too deep
into arch/m32r/platforms/*/io.c (there are plenty of more type inconsistencies
with asm-generic/io.h!), and gave up after ca. 500 lines of changes ;-)

> Signed-off-by: Arnd Bergmann 

With the below fixed:

Reviewed-by: Geert Uytterhoeven 

> --- a/arch/m32r/include/asm/io.h
> +++ b/arch/m32r/include/asm/io.h
> @@ -108,9 +108,9 @@ static inline unsigned short _readw(unsigned long addr)
> return *(volatile unsigned short __force *)addr;
>  }
>
> -static inline unsigned long _readl(unsigned long addr)
> +static inline unsigned int _readl(unsigned long addr)
>  {
> -   return *(volatile unsigned long __force *)addr;
> +   return *(volatile unsigned int __force *)addr;
>  }
>
>  static inline void _writeb(unsigned char b, unsigned long addr)
> @@ -123,9 +123,9 @@ static inline void _writew(unsigned short w, unsigned 
> long addr)
> *(volatile unsigned short __force *)addr = w;
>  }
>
> -static inline void _writel(unsigned long l, unsigned long addr)
> +static inline void _writel(unsigned long l, unsigned int addr)

Don't you want to change the type of "l" instead of the type of "addr"?

>  {
> -   *(volatile unsigned long __force *)addr = l;
> +   *(volatile unsigned int __force *)addr = l;
>  }

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds