[Skiboot] [PATCH 1/3] hdata: Don't use printf

[Vasant Hegde]

On 05/15/2017 01:50 PM, Oliver O'Halloran wrote:
> Using printf() here results in the output going into the stdout stream
> when running the hdata_to_dt test. This results in an invalid dtb output
> so lets not do that.
>
> Signed-off-by: Oliver O'Halloran <oohall at gmail.com>
> ---
>  hdata/iohub.c | 3 ++-
>  hdata/paca.c  | 4 ++--
>  2 files changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/hdata/iohub.c b/hdata/iohub.c
> index 0e7ef65f9101..c0b9bf802efb 100644
> --- a/hdata/iohub.c
> +++ b/hdata/iohub.c
> @@ -380,7 +380,8 @@ static struct dt_node *io_add_p9(const struct cechub_io_hub *hub,
>  		return NULL;
>  	}
>
> -	printf("IOHUB: PHB4 active bridge mask %x\n", (u32) hub->fab_br0_pdt);
> +	prlog(PR_DEBUG, "IOHUB: PHB4 active bridge mask %x\n",
> +		(u32) hub->fab_br0_pdt);
>
>  	/* Create PBCQs */
>  	io_add_phb4(hub, sp_iohubs, xscom, 0, 1, 0);
> diff --git a/hdata/paca.c b/hdata/paca.c
> index 1938d4affbeb..eaad0d4de432 100644
> --- a/hdata/paca.c
> +++ b/hdata/paca.c
> @@ -49,7 +49,7 @@ static struct dt_node *add_cpu_node(struct dt_node *cpus,
>  	no = be32_to_cpu(id->process_interrupt_line);
>
>  	ve_flags = be32_to_cpu(id->verify_exists_flags);
> -	printf("CPU[%i]: PIR=%i RES=%i %s %s(%u threads)\n",
> +	prlog(PR_INFO, "CPU[%i]: PIR=%i RES=%i %s %s(%u threads)\n",
>  	       paca_index(paca), be32_to_cpu(id->pir), no,
>  	       ve_flags & CPU_ID_PACA_RESERVED
>  	       ? "**RESERVED**" : cpu_state(ve_flags),
> @@ -251,7 +251,7 @@ static bool __paca_parse(void)
>  			okay = false;
>  		}
>
> -		printf("CPU[%i]: PIR=%i RES=%i %s\n",
> +		prlog(PR_INFO, "CPU[%i]: PIR=%i RES=%i %s\n",

May be we should make it as PR_DEBUG. But PACA is deprecated anyway.  So doesn't 
make much difference.  Otherwise patch looks good to me.

  Reviewed-by: Vasant Hegde <hegdevasant at linux.vnet.ibm.com>

-Vasant