Uncorrectable bitflips with Linux 4.12 on SAMA5D2

Uncorrectable bitflips with Linux 4.12 on SAMA5D2

[Romain Izard]

Hello,

On a custom board based on Microchip's SAMA5D2 SoC, I am encountering
ECC problems with the NAND Flash controller. The board is currently
working with Linux v4.9.x, but it does not with Linux 4.12.

Some pages on the NAND Flash memory chip of the device have bitflips.
With Linux 4.9.x, the bitflips are fixed by the PMECC code, but with
Linux 4.12, the errors are reported as uncorrectable.

>From my end, I updated the board's device tree to include the
sama5d2.dtsi as it exists in Linux 4.12, but the rest of the file is
the same as for Linux 4.9. From my understanding, it should be
sufficient as the legacy DTB binding for the NAND controller is
supported.

The relevant extract is:

        nand0: nand at 80000000 {
            pinctrl-names = "default";
            pinctrl-0 = <&pinctrl_nand_default>;
            status = "okay";
            nand-ecc-mode = "hw";
            atmel,pmecc-cap = <8>;
            atmel,pmecc-sector-size = <512>;
            /delete-property/ atmel,nand-has-dma;
        };

The attached files show the same block read through nanddump, with
Linux 4.9 and Linux 4.12

Best regards,
-- 
Romain Izard
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ad97000_nofix.dump
Type: application/octet-stream
Size: 10880 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20170717/da725c62/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ad97000_fix.dump
Type: application/octet-stream
Size: 10880 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20170717/da725c62/attachment-0003.obj>

Uncorrectable bitflips with Linux 4.12 on SAMA5D2

[Boris Brezillon]

+Ludovic, Nicolas and Alexandre

Hi Romain,

Le Mon, 17 Jul 2017 18:11:26 +0200,
Romain Izard <romain.izard.pro@gmail.com> a ?crit :

> Hello,
> 
> On a custom board based on Microchip's SAMA5D2 SoC, I am encountering
> ECC problems with the NAND Flash controller. The board is currently
> working with Linux v4.9.x, but it does not with Linux 4.12.
> 
> Some pages on the NAND Flash memory chip of the device have bitflips.
> With Linux 4.9.x, the bitflips are fixed by the PMECC code, but with
> Linux 4.12, the errors are reported as uncorrectable.
> 
> From my end, I updated the board's device tree to include the
> sama5d2.dtsi as it exists in Linux 4.12, but the rest of the file is
> the same as for Linux 4.9. From my understanding, it should be
> sufficient as the legacy DTB binding for the NAND controller is
> supported.

I know Ludovic add some problems with the NAND controller on sama5d2,
but it was with the new bindings (see the fixes here [1]).

Your problem seems a bit different since you're using the backward
compat code.

Could you dump PMECC/NFC regs when booting 4.9 and 4.12, and see if
there's a difference?

> 
> The relevant extract is:
> 
>         nand0: nand at 80000000 {
>             pinctrl-names = "default";
>             pinctrl-0 = <&pinctrl_nand_default>;
>             status = "okay";
>             nand-ecc-mode = "hw";
>             atmel,pmecc-cap = <8>;
>             atmel,pmecc-sector-size = <512>;
>             /delete-property/ atmel,nand-has-dma;
>         };
> 
> The attached files show the same block read through nanddump, with
> Linux 4.9 and Linux 4.12

Yep, the dumps show 2 identical pages except for the bitflip in ECC
block 2, but I can't tell from this dump why PMECC fails to correct
bitflips.

Really sorry that you have to debug this problem, but I don't have a
sama5d2 platform with a NAND on it and couldn't test my
modification on this platform :-/.

Thanks,

Boris

[1]https://lkml.org/lkml/2017/7/11/95

Uncorrectable bitflips with Linux 4.12 on SAMA5D2

[Boris Brezillon]

Le Mon, 17 Jul 2017 19:48:09 +0200,
Boris Brezillon <boris.brezillon@free-electrons.com> a ?crit :

> +Ludovic, Nicolas and Alexandre
> 
> Hi Romain,
> 
> Le Mon, 17 Jul 2017 18:11:26 +0200,
> Romain Izard <romain.izard.pro@gmail.com> a ?crit :
> 
> > Hello,
> > 
> > On a custom board based on Microchip's SAMA5D2 SoC, I am encountering
> > ECC problems with the NAND Flash controller. The board is currently
> > working with Linux v4.9.x, but it does not with Linux 4.12.
> > 
> > Some pages on the NAND Flash memory chip of the device have bitflips.
> > With Linux 4.9.x, the bitflips are fixed by the PMECC code, but with
> > Linux 4.12, the errors are reported as uncorrectable.
> > 
> > From my end, I updated the board's device tree to include the
> > sama5d2.dtsi as it exists in Linux 4.12, but the rest of the file is
> > the same as for Linux 4.9. From my understanding, it should be
> > sufficient as the legacy DTB binding for the NAND controller is
> > supported.  
> 
> I know Ludovic add some problems with the NAND controller on sama5d2,
> but it was with the new bindings (see the fixes here [1]).
> 
> Your problem seems a bit different since you're using the backward
> compat code.

Can you try with the following patch applied?

--->8---
>From cb95a16cdd497137872c44d5dd4e251ec73c79c1 Mon Sep 17 00:00:00 2001
From: Boris Brezillon <boris.brezillon@free-electrons.com>
Date: Mon, 17 Jul 2017 20:09:59 +0200
Subject: [PATCH] mtd: nand: atmel: Fix DT backward compatibility in pmecc.c

PMECC caps extraction from old DT bindings is broken, thus leading to
erroneous EL registers offset, which in turn make HW ECC unusable on
sama5d2 when old bindings are in use.

Passing the NAND dev node instead of the NFC node to of_match_node()
solves the problem.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Fixes: f88fc122cc34 ("mtd: nand: Cleanup/rework the atmel_nand driver")
Cc: <stable@vger.kernel.org>
---
 drivers/mtd/nand/atmel/pmecc.c | 21 ++++++---------------
 1 file changed, 6 insertions(+), 15 deletions(-)

diff --git a/drivers/mtd/nand/atmel/pmecc.c b/drivers/mtd/nand/atmel/pmecc.c
index 55a8ee5306ea..8c210a5776bc 100644
--- a/drivers/mtd/nand/atmel/pmecc.c
+++ b/drivers/mtd/nand/atmel/pmecc.c
@@ -945,6 +945,7 @@ struct atmel_pmecc *devm_atmel_pmecc_get(struct device *userdev)
 		 */
 		struct platform_device *pdev = to_platform_device(userdev);
 		const struct atmel_pmecc_caps *caps;
+		const struct of_device_id *match;
 
 		/* No PMECC engine available. */
 		if (!of_property_read_bool(userdev->of_node,
@@ -953,21 +954,11 @@ struct atmel_pmecc *devm_atmel_pmecc_get(struct device *userdev)
 
 		caps = &at91sam9g45_caps;
 
-		/*
-		 * Try to find the NFC subnode and extract the associated caps
-		 * from there.
-		 */
-		np = of_find_compatible_node(userdev->of_node, NULL,
-					     "atmel,sama5d3-nfc");
-		if (np) {
-			const struct of_device_id *match;
-
-			match = of_match_node(atmel_pmecc_legacy_match, np);
-			if (match && match->data)
-				caps = match->data;
-
-			of_node_put(np);
-		}
+		/* Find the caps associated to the NAND dev node. */
+		match = of_match_node(atmel_pmecc_legacy_match,
+				      userdev->of_node);
+		if (match && match->data)
+			caps = match->data;
 
 		pmecc = atmel_pmecc_create(pdev, caps, 1, 2);
 	}

Uncorrectable bitflips with Linux 4.12 on SAMA5D2

[Romain Izard]

Hi Boris,

2017-07-17 20:20 GMT+02:00 Boris Brezillon <boris.brezillon@free-electrons.com>:
> Le Mon, 17 Jul 2017 19:48:09 +0200,
> Boris Brezillon <boris.brezillon@free-electrons.com> a ?crit :
>
>> +Ludovic, Nicolas and Alexandre
>>
>> Hi Romain,
>>
>> Le Mon, 17 Jul 2017 18:11:26 +0200,
>> Romain Izard <romain.izard.pro@gmail.com> a ?crit :
>>
>> > Hello,
>> >
>> > On a custom board based on Microchip's SAMA5D2 SoC, I am encountering
>> > ECC problems with the NAND Flash controller. The board is currently
>> > working with Linux v4.9.x, but it does not with Linux 4.12.
>> >
>> > Some pages on the NAND Flash memory chip of the device have bitflips.
>> > With Linux 4.9.x, the bitflips are fixed by the PMECC code, but with
>> > Linux 4.12, the errors are reported as uncorrectable.
>> >
>> > From my end, I updated the board's device tree to include the
>> > sama5d2.dtsi as it exists in Linux 4.12, but the rest of the file is
>> > the same as for Linux 4.9. From my understanding, it should be
>> > sufficient as the legacy DTB binding for the NAND controller is
>> > supported.
>>
>> I know Ludovic add some problems with the NAND controller on sama5d2,
>> but it was with the new bindings (see the fixes here [1]).
>>
>> Your problem seems a bit different since you're using the backward
>> compat code.
>
> Can you try with the following patch applied?
>
> --->8---
> From cb95a16cdd497137872c44d5dd4e251ec73c79c1 Mon Sep 17 00:00:00 2001
> From: Boris Brezillon <boris.brezillon@free-electrons.com>
> Date: Mon, 17 Jul 2017 20:09:59 +0200
> Subject: [PATCH] mtd: nand: atmel: Fix DT backward compatibility in pmecc.c
>
> PMECC caps extraction from old DT bindings is broken, thus leading to
> erroneous EL registers offset, which in turn make HW ECC unusable on
> sama5d2 when old bindings are in use.
>
> Passing the NAND dev node instead of the NFC node to of_match_node()
> solves the problem.
>
> Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
> Fixes: f88fc122cc34 ("mtd: nand: Cleanup/rework the atmel_nand driver")
> Cc: <stable@vger.kernel.org>
> ---
>  drivers/mtd/nand/atmel/pmecc.c | 21 ++++++---------------
>  1 file changed, 6 insertions(+), 15 deletions(-)
>
> diff --git a/drivers/mtd/nand/atmel/pmecc.c b/drivers/mtd/nand/atmel/pmecc.c
> index 55a8ee5306ea..8c210a5776bc 100644
> --- a/drivers/mtd/nand/atmel/pmecc.c
> +++ b/drivers/mtd/nand/atmel/pmecc.c
> @@ -945,6 +945,7 @@ struct atmel_pmecc *devm_atmel_pmecc_get(struct device *userdev)
>                  */
>                 struct platform_device *pdev = to_platform_device(userdev);
>                 const struct atmel_pmecc_caps *caps;
> +               const struct of_device_id *match;
>
>                 /* No PMECC engine available. */
>                 if (!of_property_read_bool(userdev->of_node,
> @@ -953,21 +954,11 @@ struct atmel_pmecc *devm_atmel_pmecc_get(struct device *userdev)
>
>                 caps = &at91sam9g45_caps;
>
> -               /*
> -                * Try to find the NFC subnode and extract the associated caps
> -                * from there.
> -                */
> -               np = of_find_compatible_node(userdev->of_node, NULL,
> -                                            "atmel,sama5d3-nfc");
> -               if (np) {
> -                       const struct of_device_id *match;
> -
> -                       match = of_match_node(atmel_pmecc_legacy_match, np);
> -                       if (match && match->data)
> -                               caps = match->data;
> -
> -                       of_node_put(np);
> -               }
> +               /* Find the caps associated to the NAND dev node. */
> +               match = of_match_node(atmel_pmecc_legacy_match,
> +                                     userdev->of_node);
> +               if (match && match->data)
> +                       caps = match->data;
>
>                 pmecc = atmel_pmecc_create(pdev, caps, 1, 2);
>         }


The proposed patch fixes my issue. Thank you for your quick response!

You can add my:
Tested-by: Romain Izard <romain.izard.pro@gmail.com>

Best regards,
-- 
Romain Izard

Uncorrectable bitflips with Linux 4.12 on SAMA5D2

[Boris Brezillon]

On Mon, 17 Jul 2017 20:20:08 +0200
Boris Brezillon <boris.brezillon@free-electrons.com> wrote:

> Le Mon, 17 Jul 2017 19:48:09 +0200,
> Boris Brezillon <boris.brezillon@free-electrons.com> a ?crit :
> 
> > +Ludovic, Nicolas and Alexandre
> > 
> > Hi Romain,
> > 
> > Le Mon, 17 Jul 2017 18:11:26 +0200,
> > Romain Izard <romain.izard.pro@gmail.com> a ?crit :
> >   
> > > Hello,
> > > 
> > > On a custom board based on Microchip's SAMA5D2 SoC, I am encountering
> > > ECC problems with the NAND Flash controller. The board is currently
> > > working with Linux v4.9.x, but it does not with Linux 4.12.
> > > 
> > > Some pages on the NAND Flash memory chip of the device have bitflips.
> > > With Linux 4.9.x, the bitflips are fixed by the PMECC code, but with
> > > Linux 4.12, the errors are reported as uncorrectable.
> > > 
> > > From my end, I updated the board's device tree to include the
> > > sama5d2.dtsi as it exists in Linux 4.12, but the rest of the file is
> > > the same as for Linux 4.9. From my understanding, it should be
> > > sufficient as the legacy DTB binding for the NAND controller is
> > > supported.    
> > 
> > I know Ludovic add some problems with the NAND controller on sama5d2,
> > but it was with the new bindings (see the fixes here [1]).
> > 
> > Your problem seems a bit different since you're using the backward
> > compat code.  
> 
> Can you try with the following patch applied?
> 
> --->8---  
> From cb95a16cdd497137872c44d5dd4e251ec73c79c1 Mon Sep 17 00:00:00 2001
> From: Boris Brezillon <boris.brezillon@free-electrons.com>
> Date: Mon, 17 Jul 2017 20:09:59 +0200
> Subject: [PATCH] mtd: nand: atmel: Fix DT backward compatibility in pmecc.c
> 
> PMECC caps extraction from old DT bindings is broken, thus leading to
> erroneous EL registers offset, which in turn make HW ECC unusable on
> sama5d2 when old bindings are in use.
> 
> Passing the NAND dev node instead of the NFC node to of_match_node()
> solves the problem.
> 
> Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
> Fixes: f88fc122cc34 ("mtd: nand: Cleanup/rework the atmel_nand driver")
> Cc: <stable@vger.kernel.org>

Applied to nand/fixes.

> ---
>  drivers/mtd/nand/atmel/pmecc.c | 21 ++++++---------------
>  1 file changed, 6 insertions(+), 15 deletions(-)
> 
> diff --git a/drivers/mtd/nand/atmel/pmecc.c b/drivers/mtd/nand/atmel/pmecc.c
> index 55a8ee5306ea..8c210a5776bc 100644
> --- a/drivers/mtd/nand/atmel/pmecc.c
> +++ b/drivers/mtd/nand/atmel/pmecc.c
> @@ -945,6 +945,7 @@ struct atmel_pmecc *devm_atmel_pmecc_get(struct device *userdev)
>  		 */
>  		struct platform_device *pdev = to_platform_device(userdev);
>  		const struct atmel_pmecc_caps *caps;
> +		const struct of_device_id *match;
>  
>  		/* No PMECC engine available. */
>  		if (!of_property_read_bool(userdev->of_node,
> @@ -953,21 +954,11 @@ struct atmel_pmecc *devm_atmel_pmecc_get(struct device *userdev)
>  
>  		caps = &at91sam9g45_caps;
>  
> -		/*
> -		 * Try to find the NFC subnode and extract the associated caps
> -		 * from there.
> -		 */
> -		np = of_find_compatible_node(userdev->of_node, NULL,
> -					     "atmel,sama5d3-nfc");
> -		if (np) {
> -			const struct of_device_id *match;
> -
> -			match = of_match_node(atmel_pmecc_legacy_match, np);
> -			if (match && match->data)
> -				caps = match->data;
> -
> -			of_node_put(np);
> -		}
> +		/* Find the caps associated to the NAND dev node. */
> +		match = of_match_node(atmel_pmecc_legacy_match,
> +				      userdev->of_node);
> +		if (match && match->data)
> +			caps = match->data;
>  
>  		pmecc = atmel_pmecc_create(pdev, caps, 1, 2);
>  	}
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/