Re: [PATCH] MTD: make panic_write() conditional to CONFIG_MTD_OOPS

[Florian Fainelli <ffainelli@freebox.fr>]

Hello Ira,

----- "Ira W. Snyder" <iws@ovro.caltech.edu> a écrit :

> On Tue, Jan 10, 2012 at 05:51:03PM +0100, Florian Fainelli wrote:
> > The mtdoops driver is the only user of the panic_write callback
> > make all panic_write related code conditionnal to CONFIG_MTD_OOPS
> > so we can save some code. Since CONFIG_MTD_OOPS can be a module
> > use the newly introduced IS_ENABLED() macro to cope with that.
> > 
> > It is safe to define the panic_write callback as NULL when
> > CONFIG_MTD_OOPS is disabled since the mtdoops drivers check for the
> > callback being non-NULL and returns an error if this the case.
> > 
> > Signed-off-by: Florian Fainelli <ffainelli@freebox.fr>
> > ---
[snip]
> > diff --git a/drivers/mtd/onenand/onenand_base.c
> b/drivers/mtd/onenand/onenand_base.c
> > index a061bc1..9f6a2a9 100644
> > --- a/drivers/mtd/onenand/onenand_base.c
> > +++ b/drivers/mtd/onenand/onenand_base.c
> > @@ -1736,6 +1736,7 @@ static void onenand_panic_wait(struct mtd_info
> *mtd)
> >   *
> >   * Write with ECC
> >   */
> > +#if IS_ENABLED(CONFIG_MTD_OOPS)
> 
> This #if should be moved up above onenand_panic_wait() which is only
> used from onenand_panic_write().

Indeed, thanks for spotting this.

> 
> Other than that, the patch looks good to me.

Thanks.
--
Florian

> 
> Ira
> 
> >  static int onenand_panic_write(struct mtd_info *mtd, loff_t to,
> size_t len,
> >  			 size_t *retlen, const u_char *buf)
> >  {
> > @@ -1819,6 +1820,10 @@ static int onenand_panic_write(struct
> mtd_info *mtd, loff_t to, size_t len,
> >  	*retlen = written;
> >  	return ret;
> >  }
> > +#else
> > +#define onenand_panic_write	NULL
> > +#endif
> > +
> >  
> >  /**
> >   * onenand_fill_auto_oob - [INTERN] oob auto-placement transfer
> > -- 
> > 1.7.5.4
> >