Re: [PATCH] mmc, sh: use a consistent pr_ prefix

[Simon Horman <horms@verge.net.au>]

On Thu, Nov 11, 2010 at 08:57:23AM +0100, Guennadi Liakhovetski wrote:
> On Wed, 10 Nov 2010, Simon Horman wrote:
> 
> > On Wed, Nov 10, 2010 at 08:00:49PM +0900, Paul Mundt wrote:
> > > On Wed, Nov 10, 2010 at 07:27:44PM +0900, Simon Horman wrote:
> > > > Use pr_fmt to set the prefix for pr_ messages.
> > > > I believe this method is common in other
> > > > source files in the kernel tree.
> > > > 
> > > pr_fmt() is ok when you have no better options, but in this case you have
> > > the struct device working for you, so you really should be using it.
> > > 
> > > Consider the case where you have multiple MMCIF blocks, the debug
> > > messages here won't really give you much helpful information if you have
> > > no idea which block you're on.
> > > 
> > > For example..
> > > 
> > > > diff --git a/drivers/mmc/host/sh_mmcif.c b/drivers/mmc/host/sh_mmcif.c
> > > > index 3f49273..9f37f74 100644
> > > > --- a/drivers/mmc/host/sh_mmcif.c
> > > > +++ b/drivers/mmc/host/sh_mmcif.c
> > > > @@ -221,8 +222,8 @@ static int sh_mmcif_error_manage(struct sh_mmcif_host *host)
> > > >  
> > > >  	state1 = sh_mmcif_readl(host->addr, MMCIF_CE_HOST_STS1);
> > > >  	state2 = sh_mmcif_readl(host->addr, MMCIF_CE_HOST_STS2);
> > > > -	pr_debug("%s: ERR HOST_STS1 = %08x\n", DRIVER_NAME, state1);
> > > > -	pr_debug("%s: ERR HOST_STS2 = %08x\n", DRIVER_NAME, state2);
> > > > +	pr_debug("ERR HOST_STS1 = %08x\n", state1);
> > > > +	pr_debug("ERR HOST_STS2 = %08x\n", state2);
> > > >  
> > > >  	if (state1 & STS1_CMDSEQ) {
> > > >  		sh_mmcif_bitset(host, MMCIF_CE_CMD_CTRL, CMD_CTRL_BREAK);
> > > 
> > > You can just convert these to:
> > > 
> > > 	dev_dbg(&host->pd->dev, ...);
> > > 
> > > > @@ -230,7 +231,7 @@ static int sh_mmcif_error_manage(struct sh_mmcif_host *host)
> > > >  		while (1) {
> > > >  			timeout--;
> > > >  			if (timeout < 0) {
> > > > -				pr_err(DRIVER_NAME": Forceed end of " \
> > > > +				pr_err("Forceed end of " \
> > > >  					"command sequence timeout err\n");
> > > >  				return -EIO;
> > > >  			}
> > > 
> > > dev_err(), and so on.
> > > 
> > 
> > Thanks, will do.
> 
> Unfortunately I haven't seen this earlier, but one of my just submitted 
> patches already does this. Sorry.

Thanks, my patch is well and truly withdrawn now.