[U-Boot-Users] [PATCH 16/18] avr32: Fix two warnings in atmel_mci.c

[Ken.Fuchs at bench.com]

Haavard Skinnemoen wrote:

> The warnings are harmless but annoying. Let's fix them.

If the warnings are "harmless", why are you "fixing them"?

The compiler has switches to turn off warnings, if they
annoy you too much.

Does this refactoring of the code do something more than
just avoid a warning or two?  If not, I would reject it.

> Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen at atmel.com>
> ---
>  cpu/at32ap/atmel_mci.c |    6 ++++--
>  1 files changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/cpu/at32ap/atmel_mci.c b/cpu/at32ap/atmel_mci.c
> index f59dfb5..3ce9ea5 100644
> --- a/cpu/at32ap/atmel_mci.c
> +++ b/cpu/at32ap/atmel_mci.c
> @@ -182,12 +182,13 @@ static int mmc_acmd(unsigned long cmd, 
> unsigned long arg,
>  
>  static unsigned long
>  mmc_bread(int dev, unsigned long start, lbaint_t blkcnt,
> -	  unsigned long *buffer)
> +	  void *buffer)
>  {
>  	int ret, i = 0;
>  	unsigned long resp[4];
>  	unsigned long card_status, data;
>  	unsigned long wordcount;
> +	u32 *p = buffer;
>  	u32 status;
>  
>  	if (blkcnt == 0)
> @@ -225,7 +226,7 @@ mmc_bread(int dev, unsigned long start, 
> lbaint_t blkcnt,
>  			if (status & MMCI_BIT(RXRDY)) {
>  				data = mmci_readl(RDR);
>  				/* pr_debug("%x\n", data); */
> -				*buffer++ = data;
> +				*p++ = data;
>  				wordcount++;
>  			}
>  		} while(wordcount < (mmc_blkdev.blksz / 4));
> @@ -443,6 +444,7 @@ static void mci_set_data_timeout(struct 
> mmc_csd *csd)
>  
>  	dtocyc = timeout_clks;
>  	dtomul = 0;
> +	shift = 0;
>  	while (dtocyc > 15 && dtomul < 8) {
>  		dtomul++;
>  		shift = dtomul_to_shift[dtomul];
> -- 
> 1.5.5.1

Sincerely,

Ken Fuchs