Re: [PATCH v2 3/8] MTD: xway: the latched command should be persistent

[Boris Brezillon <boris.brezillon@free-electrons.com>]

On Sun, 19 Jun 2016 14:04:51 +0200
Hauke Mehrtens <hauke@hauke-m.de> wrote:

> On 06/19/2016 01:52 PM, Boris Brezillon wrote:
> > On Sat, 18 Jun 2016 21:14:07 +0200
> > Hauke Mehrtens <hauke@hauke-m.de> wrote:
> >   
> >> From: John Crispin <john@phrozen.org>
> >>
> >> If they are not persistent a different address will be used and the
> >> controller will deactivate some pins.
> >>
> >> Signed-off-by: John Crispin <john@phrozen.org>
> >> Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
> >> ---
> >>  drivers/mtd/nand/xway_nand.c | 16 ++++++++--------
> >>  1 file changed, 8 insertions(+), 8 deletions(-)
> >>
> >> diff --git a/drivers/mtd/nand/xway_nand.c b/drivers/mtd/nand/xway_nand.c
> >> index 2889e48..346781a 100644
> >> --- a/drivers/mtd/nand/xway_nand.c
> >> +++ b/drivers/mtd/nand/xway_nand.c
> >> @@ -67,6 +67,7 @@
> >>  
> >>  struct xway_nand_data {
> >>  	struct nand_chip	chip;
> >> +	u32			xway_latchcmd;
> >>  };
> >>  
> >>  static void xway_reset_chip(struct nand_chip *chip)
> >> @@ -104,22 +105,21 @@ static void xway_select_chip(struct mtd_info *mtd, int chip)
> >>  
> >>  static void xway_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl)
> >>  {
> >> -	struct nand_chip *this = mtd_to_nand(mtd);
> >> -	unsigned long nandaddr = (unsigned long) this->IO_ADDR_W;
> >> +	struct nand_chip *chip = mtd_to_nand(mtd);
> >> +	struct xway_nand_data *data = nand_get_controller_data(chip);
> >> +	unsigned long nandaddr = (unsigned long) chip->IO_ADDR_W;
> >>  	unsigned long flags;
> >>  
> >>  	if (ctrl & NAND_CTRL_CHANGE) {
> >> -		nandaddr &= ~(NAND_WRITE_CMD | NAND_WRITE_ADDR);
> >>  		if (ctrl & NAND_CLE)
> >> -			nandaddr |= NAND_WRITE_CMD;
> >> -		else
> >> -			nandaddr |= NAND_WRITE_ADDR;
> >> -		this->IO_ADDR_W = (void __iomem *) nandaddr;
> >> +			data->xway_latchcmd = NAND_WRITE_CMD;
> >> +		else if (ctrl & NAND_ALE)
> >> +			data->xway_latchcmd = NAND_WRITE_ADDR;
> >>  	}
> >>  
> >>  	if (cmd != NAND_CMD_NONE) {
> >>  		spin_lock_irqsave(&ebu_lock, flags);
> >> -		writeb(cmd, this->IO_ADDR_W);
> >> +		writeb(cmd, (void __iomem *) (nandaddr | data->xway_latchcmd));
> >>  		while ((ltq_ebu_r32(EBU_NAND_WAIT) & NAND_WAIT_WR_C) == 0)
> >>  			;
> >>  		spin_unlock_irqrestore(&ebu_lock, flags);  
> > 
> > Not sure what you're trying to do exactly. AFAICS, your controller seems
> > to handle each addr and cmd cycles independently, so all you have to do
> > is something like that:
> > 
> > 	/*
> > 	 * I'm assuming baseaddr has been initialized to the NAND
> > 	 * controller base address.
> > 	 */
> > 
> > 	if (cmd == NAND_CMD_NONE)
> > 		return;
> > 
> > 	spin_lock_irqsave(&ebu_lock, flags);
> > 	if (ctrl & NAND_CLE)
> > 		writeb(cmd, baseaddr + NAND_WRITE_CMD);
> > 	else
> > 		writeb(cmd, baseaddr + NAND_WRITE_ADDR);
> > 
> > 	while ((ltq_ebu_r32(EBU_NAND_WAIT) & NAND_WAIT_WR_C) == 0)
> > 		;
> > 	spin_unlock_irqrestore(&ebu_lock, flags);
> >   
> 
> Is NAND_CLE always set when the line to the NAND flash chip should be
> set?

Yes it is (see the nand_command() [1] and nand_command_lp() [2]
implementations).

> Then this should work and we ignore NAND_CTRL_CHANGE .
> 
> Hauke

[1]http://lxr.free-electrons.com/source/drivers/mtd/nand/nand_base.c#L588
[2]http://lxr.free-electrons.com/source/drivers/mtd/nand/nand_base.c#L691