Re: [PATCH v3 2/3] mtd: nand: mxc_nand: implement exec_op

[Miquel Raynal]

Hi Sascha,

s.hauer@xxxxxxxxxxxxxx wrote on Thu, 16 May 2024 12:25:43 +0200:

> On Thu, May 16, 2024 at 10:32:14AM +0200, Miquel Raynal wrote:
> > Hi Sascha,
> >   
> > > +static const struct nand_op_parser mxcnd_op_parser = NAND_OP_PARSER(
> > > +	NAND_OP_PARSER_PATTERN(mxcnd_do_exec_op,
> > > +			       NAND_OP_PARSER_PAT_CMD_ELEM(false),
> > > +			       NAND_OP_PARSER_PAT_ADDR_ELEM(true, 7),
> > > +			       NAND_OP_PARSER_PAT_CMD_ELEM(true),
> > > +			       NAND_OP_PARSER_PAT_WAITRDY_ELEM(true),
> > > +			       NAND_OP_PARSER_PAT_DATA_IN_ELEM(true, MAX_DATA_SIZE)),  
> > 
> > CMD, ADDR, CMD, DATA is the RNDOUT pattern. So it is now working fine?  
> 
> Yes, RNDOUT is working now.

Excellent!

> > Or did you forget to adapt the patterns to your use case?  
> 
> Although it looks like the patterns from the pl35x-nand-controller.c,
> there is one slight difference. The 'false' in the NAND_OP_PARSER_PAT_CMD_ELEM
> above has the effect that a plain NAND_OP_PARSER_PAT_DATA_IN_ELEM is
> disallowed.

I'm not sure I follow, the above pattern means: a single command cycle
is supported, no?

Miquèl