[U-Boot] [PATCH] MIPS: Add board_early_init_f() to init_sequence

[Shinya Kuribayashi]

Stefan Roese wrote:
> This patch adds the board_early_init_f() call to the MIPS init
> sequence. A weak dummy implementation is also added which can be
> overridden by a board specific version.
> 
> This will be used by the upcoming VCTH board support.
> 
> Signed-off-by: Stefan Roese <sr at denx.de>
> ---
>  lib_mips/board.c |   10 ++++++++++
>  1 files changed, 10 insertions(+), 0 deletions(-)
> 
> diff --git a/lib_mips/board.c b/lib_mips/board.c
> index 77e1cc8..01dda69 100644
> --- a/lib_mips/board.c
> +++ b/lib_mips/board.c
> @@ -70,6 +70,15 @@ static ulong mem_malloc_brk;
>   */
>  unsigned long mips_io_port_base = -1;
>  
> +int __board_early_init_f(void)
> +{
> +	/*
> +	 * Nothing to do in this dummy implementation
> +	 */
> +	return 0;
> +}
> +int board_early_init_f(void) __attribute__((weak, alias("__board_early_init_f")));
> +
>  /*
>   * The Malloc area is immediately below the monitor copy in DRAM
>   */
> @@ -167,6 +176,7 @@ static int init_baudrate (void)
>  typedef int (init_fnc_t) (void);
>  
>  init_fnc_t *init_sequence[] = {
> +	board_early_init_f,
>  	timer_init,
>  	env_init,		/* initialize environment */
>  #ifdef CONFIG_INCA_IP

This is ok.  I don't know why people prefers alias, though :-)

Acked-by: Shinya Kuribayashi <skuribay at ruby.dti.ne.jp>