Re: [PATCH v2] mtd: nand: add ->exec_op() implementation

[Miquel RAYNAL <miquel.raynal@free-electrons.com>]

Hello Antoine,

On Fri, 8 Dec 2017 08:54:50 +0100
Antoine Tenart <antoine.tenart@free-electrons.com> wrote:

> Hi Miquel,
> 
> Be careful when defining macros: avoid to name your arguments with a
> value used in the macro that is not meant to be replaced.

That is right, it could have failed the macro.

Thank you for the tip,
Miquèl

> 
> On Thu, Dec 07, 2017 at 03:54:34PM +0100, Miquel Raynal wrote:
> > +
> > +#define NAND_OP_DATA_IN(l, buf,
> > ns)					\
> > +
> > {								\
> > +		.type =
> > NAND_OP_DATA_IN_INSTR,				\
> > +		.ctx.data =
> > {						\
> > +			.len =
> > l,					\
> > +			.buf.in =
> > buf,					\  
> 
> Here.
> 
> > +			.force_8bit =
> > false,				\
> > +		},
> > \
> > +		.delay_ns =
> > ns,						\
> > +	}
> > +
> > +#define NAND_OP_DATA_OUT(l, buf,
> > ns)					\
> > +
> > {								\
> > +		.type =
> > NAND_OP_DATA_OUT_INSTR,				\
> > +		.ctx.data =
> > {						\
> > +			.len =
> > l,					\
> > +			.buf.out =
> > buf,					\  
> 
> And here.
> 
> > +			.force_8bit =
> > false,				\
> > +		},
> > \
> > +		.delay_ns =
> > ns,						\
> > +	}  
> 
> It works in your series because these macros are always called with
> the second argument being a variable named 'buf', but that's not safe.
> 
> Thanks!
> Antoine
> 



-- 
Miquel Raynal, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com