diff mbox series

[v3,37/40] mtd: spinand: Propagate ECC information to the MTD structure

Message ID 20190919193141.7865-38-miquel.raynal@bootlin.com
State Changes Requested
Delegated to: Miquel Raynal
Headers show
Series Introduce the generic ECC engine abstraction | expand

Commit Message

Miquel Raynal Sept. 19, 2019, 7:31 p.m. UTC
This is done by default in the raw NAND core (nand_base.c) but was
missing in the SPI-NAND core. Without these two lines the ecc_strength
and ecc_step_size values are not exported to the user through sysfs.

This fix depends on recent changes and should not be backported as-is.

Fixes: 7529df465248 ("mtd: nand: Add core infrastructure to support SPI NANDs")
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/mtd/nand/spi/core.c | 4 ++++
 1 file changed, 4 insertions(+)

Comments

Emil Lenngren Feb. 17, 2020, 1:27 p.m. UTC | #1
Hi,

Den tors 19 sep. 2019 kl 21:56 skrev Miquel Raynal <miquel.raynal@bootlin.com>:
>
> This is done by default in the raw NAND core (nand_base.c) but was
> missing in the SPI-NAND core. Without these two lines the ecc_strength
> and ecc_step_size values are not exported to the user through sysfs.
>
> This fix depends on recent changes and should not be backported as-is.
>
> Fixes: 7529df465248 ("mtd: nand: Add core infrastructure to support SPI NANDs")
> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
> ---
>  drivers/mtd/nand/spi/core.c | 4 ++++
>  1 file changed, 4 insertions(+)
>
> diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c
> index a521eeb0d351..36b99f68da81 100644
> --- a/drivers/mtd/nand/spi/core.c
> +++ b/drivers/mtd/nand/spi/core.c
> @@ -1135,6 +1135,10 @@ static int spinand_init(struct spinand_device *spinand)
>
>         mtd->oobavail = ret;
>
> +       /* Propagate ECC information to mtd_info */
> +       mtd->ecc_strength = nand->ecc.ctx.conf.strength;
> +       mtd->ecc_step_size = nand->ecc.ctx.conf.step_size;
> +
>         return 0;
>
>  err_cleanup_ecc_engine:
> --

If I follow the code flow correctly, it seems this is more important
than just exposing the value to the user through sysfs. The
mtd_read_oob function uses the ecc_strength and bitflip_threshold
values to decide if it should return -EUCLEAN or 0 for a read with
correctable bitflips:

if (mtd->ecc_strength == 0)
        return 0; /* device lacks ecc */
return ret_code >= mtd->bitflip_threshold ? -EUCLEAN : 0;

The -EUCLEAN error code is then used in UBI for the wear level logic,
so it would be nice if this could be fixed.
Now both ecc_strength and bitflip_threshold is 0 for me with my SPI
NAND flash, as exposed through sysfs, so I assume the wear level logic
doesn't work right now.

/Emil
Boris Brezillon Feb. 17, 2020, 1:31 p.m. UTC | #2
On Mon, 17 Feb 2020 14:27:13 +0100
Emil Lenngren <emil.lenngren@gmail.com> wrote:

> Hi,
> 
> Den tors 19 sep. 2019 kl 21:56 skrev Miquel Raynal <miquel.raynal@bootlin.com>:
> >
> > This is done by default in the raw NAND core (nand_base.c) but was
> > missing in the SPI-NAND core. Without these two lines the ecc_strength
> > and ecc_step_size values are not exported to the user through sysfs.
> >
> > This fix depends on recent changes and should not be backported as-is.
> >
> > Fixes: 7529df465248 ("mtd: nand: Add core infrastructure to support SPI NANDs")
> > Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
> > ---
> >  drivers/mtd/nand/spi/core.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> >
> > diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c
> > index a521eeb0d351..36b99f68da81 100644
> > --- a/drivers/mtd/nand/spi/core.c
> > +++ b/drivers/mtd/nand/spi/core.c
> > @@ -1135,6 +1135,10 @@ static int spinand_init(struct spinand_device *spinand)
> >
> >         mtd->oobavail = ret;
> >
> > +       /* Propagate ECC information to mtd_info */
> > +       mtd->ecc_strength = nand->ecc.ctx.conf.strength;
> > +       mtd->ecc_step_size = nand->ecc.ctx.conf.step_size;
> > +
> >         return 0;
> >
> >  err_cleanup_ecc_engine:
> > --  
> 
> If I follow the code flow correctly, it seems this is more important
> than just exposing the value to the user through sysfs. The
> mtd_read_oob function uses the ecc_strength and bitflip_threshold
> values to decide if it should return -EUCLEAN or 0 for a read with
> correctable bitflips:
> 
> if (mtd->ecc_strength == 0)
>         return 0; /* device lacks ecc */
> return ret_code >= mtd->bitflip_threshold ? -EUCLEAN : 0;
> 
> The -EUCLEAN error code is then used in UBI for the wear level logic,
> so it would be nice if this could be fixed.
> Now both ecc_strength and bitflip_threshold is 0 for me with my SPI
> NAND flash, as exposed through sysfs, so I assume the wear level logic
> doesn't work right now.

Indeed. We should probably send this fix separately and add a Cc:stable
tag.
diff mbox series

Patch

diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c
index a521eeb0d351..36b99f68da81 100644
--- a/drivers/mtd/nand/spi/core.c
+++ b/drivers/mtd/nand/spi/core.c
@@ -1135,6 +1135,10 @@  static int spinand_init(struct spinand_device *spinand)
 
 	mtd->oobavail = ret;
 
+	/* Propagate ECC information to mtd_info */
+	mtd->ecc_strength = nand->ecc.ctx.conf.strength;
+	mtd->ecc_step_size = nand->ecc.ctx.conf.step_size;
+
 	return 0;
 
 err_cleanup_ecc_engine: