* Kernel 5.10-rc1 not mounting NAND flash (Bisected to d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits"))
@ 2020-11-04 17:33 Christophe Leroy
2020-11-04 17:38 ` Miquel Raynal
0 siblings, 1 reply; 6+ messages in thread
From: Christophe Leroy @ 2020-11-04 17:33 UTC (permalink / raw)
To: Miquel Raynal; +Cc: linuxppc-dev, linux-kernel, linux-mtd
Hi Miquel,
I'm unable to boot 5.10-rc1 on my boards. I get the following error:
[ 4.125811] nand: device found, Manufacturer ID: 0xad, Chip ID: 0x76
[ 4.131992] nand: Hynix NAND 64MiB 3,3V 8-bit
[ 4.136173] nand: 64 MiB, SLC, erase size: 16 KiB, page size: 512,
OOB size: 16
[ 4.143534] ------------[ cut here ]------------
[ 4.147934] Unsupported ECC algorithm!
[ 4.152142] WARNING: CPU: 0 PID: 1 at
drivers/mtd/nand/raw/nand_base.c:5244 nand_scan_with_ids+0x1260/0x1640
...
[ 4.332052] ---[ end trace e3a36f62cae4ac56 ]---
[ 4.336882] gpio-nand: probe of c0000000.nand failed with error -22
Bisected to commit d7157ff49a5b ("mtd: rawnand: Use the ECC framework
user input parsing bits")
My first impression is that with that change, the value set in chip->ecc.algo
by gpio_nand_probe() in drivers/mtd/nand/raw/gpio.c gets overwritten
in rawnand_dt_init()
The following change fixes the problem, though I'm not sure it is the
right fix. Can you have a look ?
diff --git a/drivers/mtd/nand/raw/nand_base.c
b/drivers/mtd/nand/raw/nand_base.c
index 1f0d542d5923..aa74797cf2da 100644
--- a/drivers/mtd/nand/raw/nand_base.c
+++ b/drivers/mtd/nand/raw/nand_base.c
@@ -5032,7 +5032,8 @@ static int rawnand_dt_init(struct nand_chip *chip)
chip->ecc.engine_type = nand->ecc.defaults.engine_type;
chip->ecc.placement = nand->ecc.user_conf.placement;
- chip->ecc.algo = nand->ecc.user_conf.algo;
+ if (chip->ecc.algo == NAND_ECC_ALGO_UNKNOWN)
+ chip->ecc.algo = nand->ecc.user_conf.algo;
chip->ecc.strength = nand->ecc.user_conf.strength;
chip->ecc.size = nand->ecc.user_conf.step_size;
---
Thanks
Christophe
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: Kernel 5.10-rc1 not mounting NAND flash (Bisected to d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits")) 2020-11-04 17:33 Kernel 5.10-rc1 not mounting NAND flash (Bisected to d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits")) Christophe Leroy @ 2020-11-04 17:38 ` Miquel Raynal 2020-11-04 18:37 ` Christophe Leroy 0 siblings, 1 reply; 6+ messages in thread From: Miquel Raynal @ 2020-11-04 17:38 UTC (permalink / raw) To: Christophe Leroy; +Cc: linuxppc-dev, linux-kernel, linux-mtd Hi Christophe, Christophe Leroy <christophe.leroy@csgroup.eu> wrote on Wed, 04 Nov 2020 18:33:53 +0100: > Hi Miquel, > > I'm unable to boot 5.10-rc1 on my boards. I get the following error: > > [ 4.125811] nand: device found, Manufacturer ID: 0xad, Chip ID: 0x76 > [ 4.131992] nand: Hynix NAND 64MiB 3,3V 8-bit > [ 4.136173] nand: 64 MiB, SLC, erase size: 16 KiB, page size: 512, OOB size: 16 > [ 4.143534] ------------[ cut here ]------------ > [ 4.147934] Unsupported ECC algorithm! > [ 4.152142] WARNING: CPU: 0 PID: 1 at drivers/mtd/nand/raw/nand_base.c:5244 nand_scan_with_ids+0x1260/0x1640 > ... > [ 4.332052] ---[ end trace e3a36f62cae4ac56 ]--- > [ 4.336882] gpio-nand: probe of c0000000.nand failed with error -22 > > Bisected to commit d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits") > > My first impression is that with that change, the value set in chip->ecc.algo > by gpio_nand_probe() in drivers/mtd/nand/raw/gpio.c gets overwritten in rawnand_dt_init() > > The following change fixes the problem, though I'm not sure it is the right fix. Can you have a look ? > > diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c > index 1f0d542d5923..aa74797cf2da 100644 > --- a/drivers/mtd/nand/raw/nand_base.c > +++ b/drivers/mtd/nand/raw/nand_base.c > @@ -5032,7 +5032,8 @@ static int rawnand_dt_init(struct nand_chip *chip) > chip->ecc.engine_type = nand->ecc.defaults.engine_type; > > chip->ecc.placement = nand->ecc.user_conf.placement; > - chip->ecc.algo = nand->ecc.user_conf.algo; > + if (chip->ecc.algo == NAND_ECC_ALGO_UNKNOWN) > + chip->ecc.algo = nand->ecc.user_conf.algo; > chip->ecc.strength = nand->ecc.user_conf.strength; > chip->ecc.size = nand->ecc.user_conf.step_size; > > --- > > Thanks > Christophe Sorry for introducing this issue, I didn't had the time to send the Fixes PR yet but I think this issue has been solved already. Could you please try with a recent linux-next? Thanks, Miquèl ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Kernel 5.10-rc1 not mounting NAND flash (Bisected to d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits")) 2020-11-04 17:38 ` Miquel Raynal @ 2020-11-04 18:37 ` Christophe Leroy 2020-11-05 7:49 ` Miquel Raynal 0 siblings, 1 reply; 6+ messages in thread From: Christophe Leroy @ 2020-11-04 18:37 UTC (permalink / raw) To: Miquel Raynal; +Cc: linuxppc-dev, linux-kernel, linux-mtd Hi Miquel, Le 04/11/2020 à 18:38, Miquel Raynal a écrit : > Hi Christophe, > > Christophe Leroy <christophe.leroy@csgroup.eu> wrote on Wed, 04 Nov > 2020 18:33:53 +0100: > >> Hi Miquel, >> >> I'm unable to boot 5.10-rc1 on my boards. I get the following error: >> >> [ 4.125811] nand: device found, Manufacturer ID: 0xad, Chip ID: 0x76 >> [ 4.131992] nand: Hynix NAND 64MiB 3,3V 8-bit >> [ 4.136173] nand: 64 MiB, SLC, erase size: 16 KiB, page size: 512, OOB size: 16 >> [ 4.143534] ------------[ cut here ]------------ >> [ 4.147934] Unsupported ECC algorithm! >> [ 4.152142] WARNING: CPU: 0 PID: 1 at drivers/mtd/nand/raw/nand_base.c:5244 nand_scan_with_ids+0x1260/0x1640 >> ... >> [ 4.332052] ---[ end trace e3a36f62cae4ac56 ]--- >> [ 4.336882] gpio-nand: probe of c0000000.nand failed with error -22 >> >> Bisected to commit d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits") >> >> My first impression is that with that change, the value set in chip->ecc.algo >> by gpio_nand_probe() in drivers/mtd/nand/raw/gpio.c gets overwritten in rawnand_dt_init() >> >> The following change fixes the problem, though I'm not sure it is the right fix. Can you have a look ? >> >> diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c >> index 1f0d542d5923..aa74797cf2da 100644 >> --- a/drivers/mtd/nand/raw/nand_base.c >> +++ b/drivers/mtd/nand/raw/nand_base.c >> @@ -5032,7 +5032,8 @@ static int rawnand_dt_init(struct nand_chip *chip) >> chip->ecc.engine_type = nand->ecc.defaults.engine_type; >> >> chip->ecc.placement = nand->ecc.user_conf.placement; >> - chip->ecc.algo = nand->ecc.user_conf.algo; >> + if (chip->ecc.algo == NAND_ECC_ALGO_UNKNOWN) >> + chip->ecc.algo = nand->ecc.user_conf.algo; >> chip->ecc.strength = nand->ecc.user_conf.strength; >> chip->ecc.size = nand->ecc.user_conf.step_size; >> >> --- >> >> Thanks >> Christophe > > Sorry for introducing this issue, I didn't had the time to send the > Fixes PR yet but I think this issue has been solved already. Could > you please try with a recent linux-next? > Sorry, same problem with "Linux version 5.10.0-rc2-next-20201104" Christophe ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Kernel 5.10-rc1 not mounting NAND flash (Bisected to d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits")) 2020-11-04 18:37 ` Christophe Leroy @ 2020-11-05 7:49 ` Miquel Raynal 2020-11-05 9:06 ` Christophe Leroy 0 siblings, 1 reply; 6+ messages in thread From: Miquel Raynal @ 2020-11-05 7:49 UTC (permalink / raw) To: Christophe Leroy; +Cc: linuxppc-dev, linux-kernel, linux-mtd Hi Christophe, Christophe Leroy <christophe.leroy@csgroup.eu> wrote on Wed, 4 Nov 2020 19:37:57 +0100: > Hi Miquel, > > Le 04/11/2020 à 18:38, Miquel Raynal a écrit : > > Hi Christophe, > > > > Christophe Leroy <christophe.leroy@csgroup.eu> wrote on Wed, 04 Nov > > 2020 18:33:53 +0100: > > > >> Hi Miquel, > >> > >> I'm unable to boot 5.10-rc1 on my boards. I get the following error: > >> > >> [ 4.125811] nand: device found, Manufacturer ID: 0xad, Chip ID: 0x76 > >> [ 4.131992] nand: Hynix NAND 64MiB 3,3V 8-bit > >> [ 4.136173] nand: 64 MiB, SLC, erase size: 16 KiB, page size: 512, OOB size: 16 > >> [ 4.143534] ------------[ cut here ]------------ > >> [ 4.147934] Unsupported ECC algorithm! > >> [ 4.152142] WARNING: CPU: 0 PID: 1 at drivers/mtd/nand/raw/nand_base.c:5244 nand_scan_with_ids+0x1260/0x1640 > >> ... > >> [ 4.332052] ---[ end trace e3a36f62cae4ac56 ]--- > >> [ 4.336882] gpio-nand: probe of c0000000.nand failed with error -22 > >> > >> Bisected to commit d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits") > >> > >> My first impression is that with that change, the value set in chip->ecc.algo > >> by gpio_nand_probe() in drivers/mtd/nand/raw/gpio.c gets overwritten in rawnand_dt_init() > >> > >> The following change fixes the problem, though I'm not sure it is the right fix. Can you have a look ? > >> > >> diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c > >> index 1f0d542d5923..aa74797cf2da 100644 > >> --- a/drivers/mtd/nand/raw/nand_base.c > >> +++ b/drivers/mtd/nand/raw/nand_base.c > >> @@ -5032,7 +5032,8 @@ static int rawnand_dt_init(struct nand_chip *chip) > >> chip->ecc.engine_type = nand->ecc.defaults.engine_type; > >> > >> chip->ecc.placement = nand->ecc.user_conf.placement; > >> - chip->ecc.algo = nand->ecc.user_conf.algo; > >> + if (chip->ecc.algo == NAND_ECC_ALGO_UNKNOWN) > >> + chip->ecc.algo = nand->ecc.user_conf.algo; > >> chip->ecc.strength = nand->ecc.user_conf.strength; > >> chip->ecc.size = nand->ecc.user_conf.step_size; > >> > >> --- > >> > >> Thanks > >> Christophe > > > > Sorry for introducing this issue, I didn't had the time to send the > > Fixes PR yet but I think this issue has been solved already. Could > > you please try with a recent linux-next? > > > > Sorry, same problem with "Linux version 5.10.0-rc2-next-20201104" Can you please give this patch a try, please? ---8<--- Author: Miquel Raynal <miquel.raynal@bootlin.com> Date: Thu Nov 5 08:44:48 2020 +0100 mtd: rawnand: gpio: Move the ECC initialization to ->attach_chip() While forcing a Hamming software ECC looks clearly wrong, let's just fix the situation for now and move these lines to the ->attach_chip() hook which gets executed after the user input parsing and NAND chip discovery. Fixes: d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits") Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com> diff --git a/drivers/mtd/nand/raw/gpio.c b/drivers/mtd/nand/raw/gpio.c index 3bd847ccc3f3..6feab847f5e0 100644 --- a/drivers/mtd/nand/raw/gpio.c +++ b/drivers/mtd/nand/raw/gpio.c @@ -161,8 +161,15 @@ static int gpio_nand_exec_op(struct nand_chip *chip, return ret; } +static int gpio_nand_attach_chip(struct nand_chip *chip) +{ + chip->ecc.mode = NAND_ECC_SOFT; + chip->ecc.algo = NAND_ECC_HAMMING; +} + static const struct nand_controller_ops gpio_nand_ops = { .exec_op = gpio_nand_exec_op, + .attach_chip = gpio_nand_attach_chip, }; #ifdef CONFIG_OF @@ -342,8 +349,6 @@ static int gpio_nand_probe(struct platform_device *pdev) gpiomtd->base.ops = &gpio_nand_ops; nand_set_flash_node(chip, pdev->dev.of_node); - chip->ecc.mode = NAND_ECC_SOFT; - chip->ecc.algo = NAND_ECC_HAMMING; chip->options = gpiomtd->plat.options; chip->controller = &gpiomtd->base; ^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: Kernel 5.10-rc1 not mounting NAND flash (Bisected to d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits")) 2020-11-05 7:49 ` Miquel Raynal @ 2020-11-05 9:06 ` Christophe Leroy 2020-11-05 9:13 ` Miquel Raynal 0 siblings, 1 reply; 6+ messages in thread From: Christophe Leroy @ 2020-11-05 9:06 UTC (permalink / raw) To: Miquel Raynal; +Cc: linuxppc-dev, linux-kernel, linux-mtd Quoting Miquel Raynal <miquel.raynal@bootlin.com>: > Hi Christophe, > > Christophe Leroy <christophe.leroy@csgroup.eu> wrote on Wed, 4 Nov 2020 > 19:37:57 +0100: > >> Hi Miquel, >> >> Le 04/11/2020 à 18:38, Miquel Raynal a écrit : >> > Hi Christophe, >> > >> > Christophe Leroy <christophe.leroy@csgroup.eu> wrote on Wed, 04 Nov >> > 2020 18:33:53 +0100: >> > >> >> Hi Miquel, >> >> >> >> I'm unable to boot 5.10-rc1 on my boards. I get the following error: >> >> >> >> [ 4.125811] nand: device found, Manufacturer ID: 0xad, Chip ID: 0x76 >> >> [ 4.131992] nand: Hynix NAND 64MiB 3,3V 8-bit >> >> [ 4.136173] nand: 64 MiB, SLC, erase size: 16 KiB, page size: >> 512, OOB size: 16 >> >> [ 4.143534] ------------[ cut here ]------------ >> >> [ 4.147934] Unsupported ECC algorithm! >> >> [ 4.152142] WARNING: CPU: 0 PID: 1 at >> drivers/mtd/nand/raw/nand_base.c:5244 >> nand_scan_with_ids+0x1260/0x1640 >> >> ... >> >> [ 4.332052] ---[ end trace e3a36f62cae4ac56 ]--- >> >> [ 4.336882] gpio-nand: probe of c0000000.nand failed with error -22 >> >> >> >> Bisected to commit d7157ff49a5b ("mtd: rawnand: Use the ECC >> framework user input parsing bits") >> >> >> >> My first impression is that with that change, the value set in >> chip->ecc.algo >> >> by gpio_nand_probe() in drivers/mtd/nand/raw/gpio.c gets >> overwritten in rawnand_dt_init() >> >> >> >> The following change fixes the problem, though I'm not sure it >> is the right fix. Can you have a look ? >> >> >> >> diff --git a/drivers/mtd/nand/raw/nand_base.c >> b/drivers/mtd/nand/raw/nand_base.c >> >> index 1f0d542d5923..aa74797cf2da 100644 >> >> --- a/drivers/mtd/nand/raw/nand_base.c >> >> +++ b/drivers/mtd/nand/raw/nand_base.c >> >> @@ -5032,7 +5032,8 @@ static int rawnand_dt_init(struct nand_chip *chip) >> >> chip->ecc.engine_type = nand->ecc.defaults.engine_type; >> >> >> >> chip->ecc.placement = nand->ecc.user_conf.placement; >> >> - chip->ecc.algo = nand->ecc.user_conf.algo; >> >> + if (chip->ecc.algo == NAND_ECC_ALGO_UNKNOWN) >> >> + chip->ecc.algo = nand->ecc.user_conf.algo; >> >> chip->ecc.strength = nand->ecc.user_conf.strength; >> >> chip->ecc.size = nand->ecc.user_conf.step_size; >> >> >> >> --- >> >> >> >> Thanks >> >> Christophe >> > >> > Sorry for introducing this issue, I didn't had the time to send the >> > Fixes PR yet but I think this issue has been solved already. Could >> > you please try with a recent linux-next? >> > >> >> Sorry, same problem with "Linux version 5.10.0-rc2-next-20201104" > > Can you please give this patch a try, please? > > ---8<--- > > Author: Miquel Raynal <miquel.raynal@bootlin.com> > Date: Thu Nov 5 08:44:48 2020 +0100 > > mtd: rawnand: gpio: Move the ECC initialization to ->attach_chip() > > While forcing a Hamming software ECC looks clearly wrong, let's just > fix the situation for now and move these lines to the ->attach_chip() > hook which gets executed after the user input parsing and NAND chip > discovery. > > Fixes: d7157ff49a5b ("mtd: rawnand: Use the ECC framework user > input parsing bits") > Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com> > > diff --git a/drivers/mtd/nand/raw/gpio.c b/drivers/mtd/nand/raw/gpio.c > index 3bd847ccc3f3..6feab847f5e0 100644 > --- a/drivers/mtd/nand/raw/gpio.c > +++ b/drivers/mtd/nand/raw/gpio.c > @@ -161,8 +161,15 @@ static int gpio_nand_exec_op(struct nand_chip *chip, > return ret; > } > > +static int gpio_nand_attach_chip(struct nand_chip *chip) > +{ > + chip->ecc.mode = NAND_ECC_SOFT; > + chip->ecc.algo = NAND_ECC_HAMMING; > +} > + > static const struct nand_controller_ops gpio_nand_ops = { > .exec_op = gpio_nand_exec_op, > + .attach_chip = gpio_nand_attach_chip, > }; > > #ifdef CONFIG_OF > @@ -342,8 +349,6 @@ static int gpio_nand_probe(struct platform_device *pdev) > gpiomtd->base.ops = &gpio_nand_ops; > > nand_set_flash_node(chip, pdev->dev.of_node); > - chip->ecc.mode = NAND_ECC_SOFT; > - chip->ecc.algo = NAND_ECC_HAMMING; > chip->options = gpiomtd->plat.options; > chip->controller = &gpiomtd->base; Works with the following: diff --git a/drivers/mtd/nand/raw/gpio.c b/drivers/mtd/nand/raw/gpio.c index 4ec0a1e10867..66d3f1eb788c 100644 --- a/drivers/mtd/nand/raw/gpio.c +++ b/drivers/mtd/nand/raw/gpio.c @@ -161,8 +161,17 @@ static int gpio_nand_exec_op(struct nand_chip *chip, return ret; } +static int gpio_nand_attach_chip(struct nand_chip *chip) +{ + chip->ecc.engine_type = NAND_ECC_ENGINE_TYPE_SOFT; + chip->ecc.algo = NAND_ECC_ALGO_HAMMING; + + return 0; +} + static const struct nand_controller_ops gpio_nand_ops = { .exec_op = gpio_nand_exec_op, + .attach_chip = gpio_nand_attach_chip, }; #ifdef CONFIG_OF @@ -342,8 +351,6 @@ static int gpio_nand_probe(struct platform_device *pdev) gpiomtd->base.ops = &gpio_nand_ops; nand_set_flash_node(chip, pdev->dev.of_node); - chip->ecc.engine_type = NAND_ECC_ENGINE_TYPE_SOFT; - chip->ecc.algo = NAND_ECC_ALGO_HAMMING; chip->options = gpiomtd->plat.options; chip->controller = &gpiomtd->base; --- Christophe ^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: Kernel 5.10-rc1 not mounting NAND flash (Bisected to d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits")) 2020-11-05 9:06 ` Christophe Leroy @ 2020-11-05 9:13 ` Miquel Raynal 0 siblings, 0 replies; 6+ messages in thread From: Miquel Raynal @ 2020-11-05 9:13 UTC (permalink / raw) To: Christophe Leroy; +Cc: linuxppc-dev, linux-kernel, linux-mtd Hi Christophe, Christophe Leroy <christophe.leroy@csgroup.eu> wrote on Thu, 05 Nov 2020 10:06:51 +0100: > Quoting Miquel Raynal <miquel.raynal@bootlin.com>: > > > Hi Christophe, > > > > Christophe Leroy <christophe.leroy@csgroup.eu> wrote on Wed, 4 Nov 2020 > > 19:37:57 +0100: > > > >> Hi Miquel, > >> > >> Le 04/11/2020 à 18:38, Miquel Raynal a écrit : > >> > Hi Christophe, > >> > > >> > Christophe Leroy <christophe.leroy@csgroup.eu> wrote on Wed, 04 Nov > >> > 2020 18:33:53 +0100: > >> > > >> >> Hi Miquel, > >> >> > >> >> I'm unable to boot 5.10-rc1 on my boards. I get the following error: > >> >> > >> >> [ 4.125811] nand: device found, Manufacturer ID: 0xad, Chip ID: 0x > 76 > >> >> [ 4.131992] nand: Hynix NAND 64MiB 3,3V 8-bit > >> >> [ 4.136173] nand: 64 MiB, SLC, erase size: 16 KiB, page size:>> 512, OOB size: 16 > >> >> [ 4.143534] ------------[ cut here ]------------ > >> >> [ 4.147934] Unsupported ECC algorithm! > >> >> [ 4.152142] WARNING: CPU: 0 PID: 1 at >> drivers/mtd/nand/raw/nand_base.c:5244 >> nand_scan_with_ids+0x1260/0x1640 > >> >> ... > >> >> [ 4.332052] ---[ end trace e3a36f62cae4ac56 ]--- > >> >> [ 4.336882] gpio-nand: probe of c0000000.nand failed with error -2 > 2 > >> >> > >> >> Bisected to commit d7157ff49a5b ("mtd: rawnand: Use the ECC >> framework user input parsing bits") > >> >> > >> >> My first impression is that with that change, the value set in >> chip->ecc.algo > >> >> by gpio_nand_probe() in drivers/mtd/nand/raw/gpio.c gets >> overwritten in rawnand_dt_init() > >> >> > >> >> The following change fixes the problem, though I'm not sure it >> is the right fix. Can you have a look ? > >> >> > >> >> diff --git a/drivers/mtd/nand/raw/nand_base.c >> b/drivers/mtd/nand/raw/nand_base.c > >> >> index 1f0d542d5923..aa74797cf2da 100644 > >> >> --- a/drivers/mtd/nand/raw/nand_base.c > >> >> +++ b/drivers/mtd/nand/raw/nand_base.c > >> >> @@ -5032,7 +5032,8 @@ static int rawnand_dt_init(struct nand_chip *ch > ip) > >> >> chip->ecc.engine_type = nand->ecc.defaults.engine_type; > >> >> > >> >> chip->ecc.placement = nand->ecc.user_conf.placement; > >> >> - chip->ecc.algo = nand->ecc.user_conf.algo; > >> >> + if (chip->ecc.algo == NAND_ECC_ALGO_UNKNOWN) > >> >> + chip->ecc.algo = nand->ecc.user_conf.algo; > >> >> chip->ecc.strength = nand->ecc.user_conf.strength; > >> >> chip->ecc.size = nand->ecc.user_conf.step_size; > >> >> > >> >> --- > >> >> > >> >> Thanks > >> >> Christophe > >> > > >> > Sorry for introducing this issue, I didn't had the time to send the > >> > Fixes PR yet but I think this issue has been solved already. Could > >> > you please try with a recent linux-next? > >> > > >> > >> Sorry, same problem with "Linux version 5.10.0-rc2-next-20201104" > > > > Can you please give this patch a try, please? > > > > ---8<--- > > > > Author: Miquel Raynal <miquel.raynal@bootlin.com> > > Date: Thu Nov 5 08:44:48 2020 +0100 > > > > mtd: rawnand: gpio: Move the ECC initialization to ->attach_chip() > > > > While forcing a Hamming software ECC looks clearly wrong, let's just > > fix the situation for now and move these lines to the ->attach_chip() > > hook which gets executed after the user input parsing and NAND chip > > discovery. > > > > Fixes: d7157ff49a5b ("mtd: rawnand: Use the ECC framework user > input parsing bits") > > Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com> > > > > diff --git a/drivers/mtd/nand/raw/gpio.c b/drivers/mtd/nand/raw/gpio.c > > index 3bd847ccc3f3..6feab847f5e0 100644 > > --- a/drivers/mtd/nand/raw/gpio.c > > +++ b/drivers/mtd/nand/raw/gpio.c > > @@ -161,8 +161,15 @@ static int gpio_nand_exec_op(struct nand_chip *chip, > > return ret; > > } > > > > +static int gpio_nand_attach_chip(struct nand_chip *chip) > > +{ > > + chip->ecc.mode = NAND_ECC_SOFT; > > + chip->ecc.algo = NAND_ECC_HAMMING; > > +} > > + > > static const struct nand_controller_ops gpio_nand_ops = { > > .exec_op = gpio_nand_exec_op, > > + .attach_chip = gpio_nand_attach_chip, > > }; > > > > #ifdef CONFIG_OF > > @@ -342,8 +349,6 @@ static int gpio_nand_probe(struct platform_device *pd > ev) > > gpiomtd->base.ops = &gpio_nand_ops; > > > > nand_set_flash_node(chip, pdev->dev.of_node); > > - chip->ecc.mode = NAND_ECC_SOFT; > > - chip->ecc.algo = NAND_ECC_HAMMING; > > chip->options = gpiomtd->plat.options; > > chip->controller = &gpiomtd->base; > > > Works with the following: > > diff --git a/drivers/mtd/nand/raw/gpio.c b/drivers/mtd/nand/raw/gpio.c > index 4ec0a1e10867..66d3f1eb788c 100644 > --- a/drivers/mtd/nand/raw/gpio.c > +++ b/drivers/mtd/nand/raw/gpio.c > @@ -161,8 +161,17 @@ static int gpio_nand_exec_op(struct nand_chip *chip, > return ret; > } > > +static int gpio_nand_attach_chip(struct nand_chip *chip) > +{ > + chip->ecc.engine_type = NAND_ECC_ENGINE_TYPE_SOFT; > + chip->ecc.algo = NAND_ECC_ALGO_HAMMING; > + > + return 0; > +} Yup indeed it was not even compile tested. Good to know, I'll check the other drivers and send the patch soon. Thanks, Miquèl ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2020-11-05 9:13 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-11-04 17:33 Kernel 5.10-rc1 not mounting NAND flash (Bisected to d7157ff49a5b ("mtd: rawnand: Use the ECC framework user input parsing bits")) Christophe Leroy
2020-11-04 17:38 ` Miquel Raynal
2020-11-04 18:37 ` Christophe Leroy
2020-11-05 7:49 ` Miquel Raynal
2020-11-05 9:06 ` Christophe Leroy
2020-11-05 9:13 ` Miquel Raynal
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).