Re: [PATCH] ipmi: fix slave-addr property add to ipmi platform device

[Corey Minyard <minyard@acm.org>]

On Wed, Apr 24, 2019 at 10:34:14AM +0000, Kamlakant Patel wrote:
> Remove check for p->slave_addr while adding 'slave-addr' property entry
> since it can be 0.
> Currently, it throws following error:
> "ipmi_ssif dmi-ipmi-ssif.0: ipmi_ssif: device has no slave-addr
> property"
> 
> This patch fixes the issue.

This would cause things that don't supply slave addresses to get
the wrong address, they will get 0x00 instead of 0x20, like they
should.

The right thing is to remote the warning, I guess.

-corey

> 
> Signed-off-by: Kamlakant Patel <kamlakantp@marvell.com>
> ---
>  drivers/char/ipmi/ipmi_plat_data.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/drivers/char/ipmi/ipmi_plat_data.c b/drivers/char/ipmi/ipmi_plat_data.c
> index 8f0ca2a..2b62104 100644
> --- a/drivers/char/ipmi/ipmi_plat_data.c
> +++ b/drivers/char/ipmi/ipmi_plat_data.c
> @@ -34,8 +34,7 @@ struct platform_device *ipmi_platform_add(const char *name, unsigned int inst,
>  		p->regspacing = p->regsize;
>  
>  	pr[pidx++] = PROPERTY_ENTRY_U8("ipmi-type", p->type);
> -	if (p->slave_addr)
> -		pr[pidx++] = PROPERTY_ENTRY_U8("slave-addr", p->slave_addr);
> +	pr[pidx++] = PROPERTY_ENTRY_U8("slave-addr", p->slave_addr);
>  	pr[pidx++] = PROPERTY_ENTRY_U8("addr-source", p->addr_source);
>  	if (p->regshift)
>  		pr[pidx++] = PROPERTY_ENTRY_U8("reg-shift", p->regshift);
> -- 
> 1.8.3.1
>