On Thu, 22 Feb 2018 20:15:58 +0530
Shreeya Patel <shreeya.patel23...@gmail.com> wrote:

> On Thu, 2018-02-22 at 15:30 +0100, Richard Weinberger wrote:
> > Am Donnerstag, 22. Februar 2018, 14:10:14 CET schrieb Shreeya Patel:  
> > > 
> > > @@ -438,7 +438,7 @@ static void __init doc2000_count_chips(struct
> > > mtd_info
> > > *mtd) break;
> > >   }
> > >   doc->chips_per_floor = i;
> > > - printk(KERN_DEBUG "Detected %d chips per floor.\n", i);
> > > + pr_debug("Detected %d chips per floor.\n", i);  
> > <bike-shedding>
> > 
> > You are aware of the fact that printk(KERN_DEBUG and pr_debug() are
> > not 
> > equivalent?
> > Not a big deal, just want to make sure that you understand what you
> > are doing.   
> 
> Actually pr_debug() shouldn't be used here.
> 
> We usually don't prefer pr_debug for drivers, instead we use dev_dbg.
> But as Boris told me that it will be better to first having changed all
> printks to pr_*() macro and then think about dev_*() macros.

BTW, I see a bunch of remaining printks when I do 'git grep printk
drivers/mtd/nand', basically all those that don't have a
KERN_<LOGLEVEL> in it. I know your commit log explicitly says that only
printks embedding a log level are converted to pr_xxx(), but it's
probably worth patching those printks as well (choosing the appropriate
log level between info, err and warn and use pr_xxx() helpers).
This should be done in a separate patch of course.

> 
> > :-)
> > 
> > </bike-shedding>
> > 
> > Thanks,
> > //richard  



-- 
Boris Brezillon, Bootlin (formerly Free Electrons)
Embedded Linux and Kernel engineering
https://bootlin.com

Reply via email to