| Message ID | 20240325144148.3738195-7-avromanov@salutedevices.com |
|---|---|
| State | New |
| Delegated to: | Dario Binacchi |
| Headers | show |
| Series | Introduce UBI block device | expand |
On 25.03.24 15:41, Alexey Romanov wrote: > UBI block is virtual block device, which is an abstraction > over MTD layer. Therefore it is logical to use it in combination > with MTD drivers. > > Signed-off-by: Alexey Romanov <avromanov@salutedevices.com> Reviewed-by: Frieder Schrempf <frieder.schrempf@kontron.de> > --- > drivers/mtd/nand/spi/core.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > > diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c > index dd880adf31..c47f6c1b46 100644 > --- a/drivers/mtd/nand/spi/core.c > +++ b/drivers/mtd/nand/spi/core.c > @@ -27,6 +27,7 @@ > #include <watchdog.h> > #include <spi.h> > #include <spi-mem.h> > +#include <ubi_uboot.h> > #include <dm/device_compat.h> > #include <dm/devres.h> > #include <dm/uclass.h> > @@ -1182,8 +1183,13 @@ static int spinand_bind(struct udevice *dev) > { > if (blk_enabled()) { > struct spinand_plat *plat = dev_get_plat(dev); > + int ret; > + > + ret = mtd_bind(dev, &plat->mtd); > + if (ret) > + return ret; > > - return mtd_bind(dev, &plat->mtd); > + return ubi_bind(dev); > } > > return 0;
Hello! On Mon, Mar 25, 2024 at 10:46 PM Alexey Romanov <avromanov@salutedevices.com> wrote: > > UBI block is virtual block device, which is an abstraction > over MTD layer. Therefore it is logical to use it in combination > with MTD drivers. > > Signed-off-by: Alexey Romanov <avromanov@salutedevices.com> > --- > drivers/mtd/nand/spi/core.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > > diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c > index dd880adf31..c47f6c1b46 100644 > --- a/drivers/mtd/nand/spi/core.c > +++ b/drivers/mtd/nand/spi/core.c > @@ -27,6 +27,7 @@ > #include <watchdog.h> > #include <spi.h> > #include <spi-mem.h> > +#include <ubi_uboot.h> > #include <dm/device_compat.h> > #include <dm/devres.h> > #include <dm/uclass.h> > @@ -1182,8 +1183,13 @@ static int spinand_bind(struct udevice *dev) > { > if (blk_enabled()) { > struct spinand_plat *plat = dev_get_plat(dev); > + int ret; > + > + ret = mtd_bind(dev, &plat->mtd); > + if (ret) > + return ret; > > - return mtd_bind(dev, &plat->mtd); > + return ubi_bind(dev); Is this expecting the entire SPI-NAND covered by a single UBI partition? It's almost never this case on real hardware. For SPI-NAND booted SoCs, the first few blocks always store the first stage bootloader raw, because bootrom knows nothing about UBI. > } > > return 0; > -- > 2.34.1 >
Hi, On Thu, Apr 04, 2024 at 11:23:47AM +0800, Chuanhong Guo wrote: > Hello! > > On Mon, Mar 25, 2024 at 10:46 PM Alexey Romanov > <avromanov@salutedevices.com> wrote: > > > > UBI block is virtual block device, which is an abstraction > > over MTD layer. Therefore it is logical to use it in combination > > with MTD drivers. > > > > Signed-off-by: Alexey Romanov <avromanov@salutedevices.com> > > --- > > drivers/mtd/nand/spi/core.c | 8 +++++++- > > 1 file changed, 7 insertions(+), 1 deletion(-) > > > > diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c > > index dd880adf31..c47f6c1b46 100644 > > --- a/drivers/mtd/nand/spi/core.c > > +++ b/drivers/mtd/nand/spi/core.c > > @@ -27,6 +27,7 @@ > > #include <watchdog.h> > > #include <spi.h> > > #include <spi-mem.h> > > +#include <ubi_uboot.h> > > #include <dm/device_compat.h> > > #include <dm/devres.h> > > #include <dm/uclass.h> > > @@ -1182,8 +1183,13 @@ static int spinand_bind(struct udevice *dev) > > { > > if (blk_enabled()) { > > struct spinand_plat *plat = dev_get_plat(dev); > > + int ret; > > + > > + ret = mtd_bind(dev, &plat->mtd); > > + if (ret) > > + return ret; > > > > - return mtd_bind(dev, &plat->mtd); > > + return ubi_bind(dev); > > Is this expecting the entire SPI-NAND covered by a single UBI partition? Why? Nah. ubi_bind() just create block device and bind it to SPI-NAND device. When working with this block device user must specify, which SPI-NAND partition UBI is located on. > It's almost never this case on real hardware. For SPI-NAND booted > SoCs, the first few blocks always store the first stage bootloader > raw, because bootrom knows nothing about UBI. > > > } > > > > return 0; > > -- > > 2.34.1 > > > > > -- > Regards, > Chuanhong Guo
Hi! On Thu, Apr 4, 2024 at 4:02 PM Alexey Romanov <avromanov@salutedevices.com> wrote: > > > +#include <ubi_uboot.h> > > > #include <dm/device_compat.h> > > > #include <dm/devres.h> > > > #include <dm/uclass.h> > > > @@ -1182,8 +1183,13 @@ static int spinand_bind(struct udevice *dev) > > > { > > > if (blk_enabled()) { > > > struct spinand_plat *plat = dev_get_plat(dev); > > > + int ret; > > > + > > > + ret = mtd_bind(dev, &plat->mtd); > > > + if (ret) > > > + return ret; > > > > > > - return mtd_bind(dev, &plat->mtd); > > > + return ubi_bind(dev); > > > > Is this expecting the entire SPI-NAND covered by a single UBI partition? > > Why? Nah. I can't find the code it's patching in my outdated local tree and made a wrong guess. Sorry for my ignorance. > ubi_bind() just create block device and bind it to SPI-NAND device. > When working with this block device user must specify, which SPI-NAND > partition UBI is located on. > And thanks for the explanation!
diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c index dd880adf31..c47f6c1b46 100644 --- a/drivers/mtd/nand/spi/core.c +++ b/drivers/mtd/nand/spi/core.c @@ -27,6 +27,7 @@ #include <watchdog.h> #include <spi.h> #include <spi-mem.h> +#include <ubi_uboot.h> #include <dm/device_compat.h> #include <dm/devres.h> #include <dm/uclass.h> @@ -1182,8 +1183,13 @@ static int spinand_bind(struct udevice *dev) { if (blk_enabled()) { struct spinand_plat *plat = dev_get_plat(dev); + int ret; + + ret = mtd_bind(dev, &plat->mtd); + if (ret) + return ret; - return mtd_bind(dev, &plat->mtd); + return ubi_bind(dev); } return 0;
UBI block is virtual block device, which is an abstraction over MTD layer. Therefore it is logical to use it in combination with MTD drivers. Signed-off-by: Alexey Romanov <avromanov@salutedevices.com> --- drivers/mtd/nand/spi/core.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)