[PATCH] mtd: rawnand: fsmc: Make sure we wait tWB before polling the STATUS reg

[PATCH] mtd: rawnand: fsmc: Make sure we wait tWB before polling the STATUS reg

[Boris Brezillon]

NAND chips require a bit of time to take the NAND operation into account
and set the BUSY bit in the STATUS reg. Make sure we don't poll the
STATUS reg too early.

Fixes: 4da712e70294 ("mtd: nand: fsmc: use ->exec_op()")
Signed-off-by: Boris Brezillon <boris.brezillon@bootlin.com>
---
 drivers/mtd/nand/raw/fsmc_nand.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/mtd/nand/raw/fsmc_nand.c b/drivers/mtd/nand/raw/fsmc_nand.c
index 28c48dcc514e..0960665858b7 100644
--- a/drivers/mtd/nand/raw/fsmc_nand.c
+++ b/drivers/mtd/nand/raw/fsmc_nand.c
@@ -18,6 +18,7 @@
 
 #include <linux/clk.h>
 #include <linux/completion.h>
+#include <linux/delay.h>
 #include <linux/dmaengine.h>
 #include <linux/dma-direction.h>
 #include <linux/dma-mapping.h>
@@ -695,6 +696,13 @@ static int fsmc_exec_op(struct nand_chip *chip, const struct nand_operation *op,
 			pr_debug("  ->WAITRDY  [max %d ms]\n",
 				 instr->ctx.waitrdy.timeout_ms);
 
+			/*
+			 * Make sure we wait tWB before polling the STATUS
+			 * register.
+			 */
+			if (op_id && op->instrs[op_id - 1].delay_ns)
+				ndelay(op->instrs[op_id - 1].delay_ns);
+
 			ret = nand_soft_waitrdy(chip,
 						instr->ctx.waitrdy.timeout_ms);
 			break;
-- 
2.14.1

Re: [PATCH] mtd: rawnand: fsmc: Make sure we wait tWB before polling the STATUS reg

[Miquel Raynal]

Hi Boris,

On Thu,  3 May 2018 09:45:44 +0200, Boris Brezillon
<boris.brezillon@bootlin.com> wrote:

> NAND chips require a bit of time to take the NAND operation into account
> and set the BUSY bit in the STATUS reg. Make sure we don't poll the
> STATUS reg too early.
> 
> Fixes: 4da712e70294 ("mtd: nand: fsmc: use ->exec_op()")
> Signed-off-by: Boris Brezillon <boris.brezillon@bootlin.com>
> ---
>  drivers/mtd/nand/raw/fsmc_nand.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/drivers/mtd/nand/raw/fsmc_nand.c b/drivers/mtd/nand/raw/fsmc_nand.c
> index 28c48dcc514e..0960665858b7 100644
> --- a/drivers/mtd/nand/raw/fsmc_nand.c
> +++ b/drivers/mtd/nand/raw/fsmc_nand.c
> @@ -18,6 +18,7 @@
>  
>  #include <linux/clk.h>
>  #include <linux/completion.h>
> +#include <linux/delay.h>
>  #include <linux/dmaengine.h>
>  #include <linux/dma-direction.h>
>  #include <linux/dma-mapping.h>
> @@ -695,6 +696,13 @@ static int fsmc_exec_op(struct nand_chip *chip, const struct nand_operation *op,
>  			pr_debug("  ->WAITRDY  [max %d ms]\n",
>  				 instr->ctx.waitrdy.timeout_ms);
>  
> +			/*
> +			 * Make sure we wait tWB before polling the STATUS
> +			 * register.
> +			 */
> +			if (op_id && op->instrs[op_id - 1].delay_ns)
> +				ndelay(op->instrs[op_id - 1].delay_ns);
> +
>  			ret = nand_soft_waitrdy(chip,
>  						instr->ctx.waitrdy.timeout_ms);
>  			break;

I'm afraid that we encounter this exact same issue with all the drivers
using nand_soft_waitrdy() whenever the controller does not already
wait for tWB. Could we force a tWB_max delay directly inside
nand_soft_waitrdy()?

Thanks,
Miquèl

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

Re: [PATCH] mtd: rawnand: fsmc: Make sure we wait tWB before polling the STATUS reg

[Boris Brezillon]

On Fri, 4 May 2018 11:45:10 +0200
Miquel Raynal <miquel.raynal@bootlin.com> wrote:

> Hi Boris,
> 
> On Thu,  3 May 2018 09:45:44 +0200, Boris Brezillon
> <boris.brezillon@bootlin.com> wrote:
> 
> > NAND chips require a bit of time to take the NAND operation into account
> > and set the BUSY bit in the STATUS reg. Make sure we don't poll the
> > STATUS reg too early.
> > 
> > Fixes: 4da712e70294 ("mtd: nand: fsmc: use ->exec_op()")
> > Signed-off-by: Boris Brezillon <boris.brezillon@bootlin.com>
> > ---
> >  drivers/mtd/nand/raw/fsmc_nand.c | 8 ++++++++
> >  1 file changed, 8 insertions(+)
> > 
> > diff --git a/drivers/mtd/nand/raw/fsmc_nand.c b/drivers/mtd/nand/raw/fsmc_nand.c
> > index 28c48dcc514e..0960665858b7 100644
> > --- a/drivers/mtd/nand/raw/fsmc_nand.c
> > +++ b/drivers/mtd/nand/raw/fsmc_nand.c
> > @@ -18,6 +18,7 @@
> >  
> >  #include <linux/clk.h>
> >  #include <linux/completion.h>
> > +#include <linux/delay.h>
> >  #include <linux/dmaengine.h>
> >  #include <linux/dma-direction.h>
> >  #include <linux/dma-mapping.h>
> > @@ -695,6 +696,13 @@ static int fsmc_exec_op(struct nand_chip *chip, const struct nand_operation *op,
> >  			pr_debug("  ->WAITRDY  [max %d ms]\n",
> >  				 instr->ctx.waitrdy.timeout_ms);
> >  
> > +			/*
> > +			 * Make sure we wait tWB before polling the STATUS
> > +			 * register.
> > +			 */
> > +			if (op_id && op->instrs[op_id - 1].delay_ns)
> > +				ndelay(op->instrs[op_id - 1].delay_ns);
> > +
> >  			ret = nand_soft_waitrdy(chip,
> >  						instr->ctx.waitrdy.timeout_ms);
> >  			break;  
> 
> I'm afraid that we encounter this exact same issue with all the drivers
> using nand_soft_waitrdy() whenever the controller does not already
> wait for tWB. Could we force a tWB_max delay directly inside
> nand_soft_waitrdy()?

Yep, I agree. I'll send a new version doing that.

Thanks,

Boris