Re: [PATCH 1/3] MIPS: Alchemy: db1200: improve PB1200 detection.

[Sergei Shtylyov <sshtylyov@mvista.com>]

Hello.

On 16-11-2011 19:42, Manuel Lauss wrote:

> The PB1200 has the CPLD located at an address which on the DB1200 is
> RAM;  reading the Board-ID sometimes results in a PB1200 being detected
> instead (especially during reboots after long uptimes).
> On the other hand, the address of the DB1200's CPLD is hosting Flash
> chips on the PB1200.  Test for the DB1200 first and additionally do a
> quick write-test to the hexleds register to make sure we're writing
> to the CPLD.

> Signed-off-by: Manuel Lauss<manuel.lauss@googlemail.com>
> ---
> Applies on top of the other patches queued for 3.3

>   arch/mips/alchemy/devboards/db1200.c |   30 ++++++++++++++++++++++--------
>   1 files changed, 22 insertions(+), 8 deletions(-)

> diff --git a/arch/mips/alchemy/devboards/db1200.c b/arch/mips/alchemy/devboards/db1200.c
> index 1181241..6721991 100644
> --- a/arch/mips/alchemy/devboards/db1200.c
> +++ b/arch/mips/alchemy/devboards/db1200.c
> @@ -66,19 +66,33 @@ static int __init detect_board(void)
>   {
>   	int bid;
>
> -	/* try the PB1200 first */
> +	/* try the DB1200 first */
> +	bcsr_init(DB1200_BCSR_PHYS_ADDR,
> +		  DB1200_BCSR_PHYS_ADDR + DB1200_BCSR_HEXLED_OFS);
> +	if (BCSR_WHOAMI_DB1200 == BCSR_WHOAMI_BOARD(bcsr_read(BCSR_WHOAMI))) {
> +		unsigned short t = bcsr_read(BCSR_HEXLEDS);
> +		bcsr_write(BCSR_HEXLEDS, ~t);
> +		if (bcsr_read(BCSR_HEXLEDS) != t) {
> +			bcsr_write(BCSR_HEXLEDS, t);
> +			return 0;
> +		}
> +	}
> +
> +	/* okay, try the PB1200 then */
>   	bcsr_init(PB1200_BCSR_PHYS_ADDR,
>   		  PB1200_BCSR_PHYS_ADDR + PB1200_BCSR_HEXLED_OFS);
>   	bid = BCSR_WHOAMI_BOARD(bcsr_read(BCSR_WHOAMI));
>   	if ((bid == BCSR_WHOAMI_PB1200_DDR1) ||
> -	    (bid == BCSR_WHOAMI_PB1200_DDR2))
> -		return 0;
> +	    (bid == BCSR_WHOAMI_PB1200_DDR2)) {
> +		unsigned short t = bcsr_read(BCSR_HEXLEDS);
> +		bcsr_write(BCSR_HEXLEDS, ~t);
> +		if (bcsr_read(BCSR_HEXLEDS) != t) {
> +			bcsr_write(BCSR_HEXLEDS, t);
> +			return 0;
> +		}

    Isn't it worth putting the repetitive code into a subroutine?

WBR, Sergei