| Message ID | 20200129155557.32191-3-yamada.masahiro@socionext.com |
|---|---|
| State | New |
| Headers | show |
| Series | [v2,1/4] mtd: rawnand: denali_dt: insert udelay() after reset deassert | expand |
Hi Masahiro, Masahiro Yamada <yamada.masahiro at socionext.com> wrote on Thu, 30 Jan 2020 00:55:56 +0900: > UCLASS_MTD is a better fit for NAND drivers. > > Make NAND_DENALI_DT depend on DM_MTD, which is needed to compile > drivers/mtd/mtd-uclass.c > > Also, make ARCH_SOCFPGA and ARCH_UNIPHIER select DM_MTD because > they use this driver. > > Signed-off-by: Masahiro Yamada <yamada.masahiro at socionext.com> > --- Reviewed-by: Miquel Raynal <miquel.raynal at bootlin.com> Thanks, Miquèl
Am 29.01.2020 um 16:55 schrieb Masahiro Yamada: > UCLASS_MTD is a better fit for NAND drivers. > > Make NAND_DENALI_DT depend on DM_MTD, which is needed to compile > drivers/mtd/mtd-uclass.c > > Also, make ARCH_SOCFPGA and ARCH_UNIPHIER select DM_MTD because > they use this driver. > > Signed-off-by: Masahiro Yamada <yamada.masahiro at socionext.com> > --- > > Changes in v2: > - new patch > > arch/arm/Kconfig | 2 ++ > drivers/mtd/nand/raw/Kconfig | 2 +- > drivers/mtd/nand/raw/denali_dt.c | 4 ++-- > 3 files changed, 5 insertions(+), 3 deletions(-) > > diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig > index 1236315168..d1c58667c7 100644 > --- a/arch/arm/Kconfig > +++ b/arch/arm/Kconfig > @@ -935,6 +935,7 @@ config ARCH_SOCFPGA > select ARM64 if TARGET_SOCFPGA_STRATIX10 || TARGET_SOCFPGA_AGILEX > select CPU_V7A if TARGET_SOCFPGA_GEN5 || TARGET_SOCFPGA_ARRIA10 > select DM > + select DM_MTD Does any socfpga actually enable the denali driver? Do we need to default to this instead of enabling it in a defconfig? If we need it, could you please change that to 'imply'? Not all configs will need this. Thanks, Simon > select DM_SERIAL > select ENABLE_ARM_SOC_BOOT0_HOOK if TARGET_SOCFPGA_GEN5 || TARGET_SOCFPGA_ARRIA10 > select OF_CONTROL > @@ -1548,6 +1549,7 @@ config ARCH_UNIPHIER > select DM_GPIO > select DM_I2C > select DM_MMC > + select DM_MTD > select DM_RESET > select DM_SERIAL > select DM_USB > diff --git a/drivers/mtd/nand/raw/Kconfig b/drivers/mtd/nand/raw/Kconfig > index 7814d84ba0..23201ca720 100644 > --- a/drivers/mtd/nand/raw/Kconfig > +++ b/drivers/mtd/nand/raw/Kconfig > @@ -116,7 +116,7 @@ config NAND_DENALI > config NAND_DENALI_DT > bool "Support Denali NAND controller as a DT device" > select NAND_DENALI > - depends on OF_CONTROL && DM > + depends on OF_CONTROL && DM_MTD > help > Enable the driver for NAND flash on platforms using a Denali NAND > controller as a DT device. > diff --git a/drivers/mtd/nand/raw/denali_dt.c b/drivers/mtd/nand/raw/denali_dt.c > index 587e480faa..759ad40e51 100644 > --- a/drivers/mtd/nand/raw/denali_dt.c > +++ b/drivers/mtd/nand/raw/denali_dt.c > @@ -160,7 +160,7 @@ static int denali_dt_probe(struct udevice *dev) > > U_BOOT_DRIVER(denali_nand_dt) = { > .name = "denali-nand-dt", > - .id = UCLASS_MISC, > + .id = UCLASS_MTD, > .of_match = denali_nand_dt_ids, > .probe = denali_dt_probe, > .priv_auto_alloc_size = sizeof(struct denali_nand_info), > @@ -171,7 +171,7 @@ void board_nand_init(void) > struct udevice *dev; > int ret; > > - ret = uclass_get_device_by_driver(UCLASS_MISC, > + ret = uclass_get_device_by_driver(UCLASS_MTD, > DM_GET_DRIVER(denali_nand_dt), > &dev); > if (ret && ret != -ENODEV) >
On 1/29/20 6:08 PM, Simon Goldschmidt wrote: > Am 29.01.2020 um 16:55 schrieb Masahiro Yamada: >> UCLASS_MTD is a better fit for NAND drivers. >> >> Make NAND_DENALI_DT depend on DM_MTD, which is needed to compile >> drivers/mtd/mtd-uclass.c >> >> Also, make ARCH_SOCFPGA and ARCH_UNIPHIER select DM_MTD because >> they use this driver. >> >> Signed-off-by: Masahiro Yamada <yamada.masahiro at socionext.com> >> --- >> >> Changes in v2: >> - new patch >> >> arch/arm/Kconfig | 2 ++ >> drivers/mtd/nand/raw/Kconfig | 2 +- >> drivers/mtd/nand/raw/denali_dt.c | 4 ++-- >> 3 files changed, 5 insertions(+), 3 deletions(-) >> >> diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig >> index 1236315168..d1c58667c7 100644 >> --- a/arch/arm/Kconfig >> +++ b/arch/arm/Kconfig >> @@ -935,6 +935,7 @@ config ARCH_SOCFPGA >> select ARM64 if TARGET_SOCFPGA_STRATIX10 || TARGET_SOCFPGA_AGILEX >> select CPU_V7A if TARGET_SOCFPGA_GEN5 || TARGET_SOCFPGA_ARRIA10 >> select DM >> + select DM_MTD > > Does any socfpga actually enable the denali driver? Do we need to > default to this instead of enabling it in a defconfig? > > If we need it, could you please change that to 'imply'? Not all configs > will need this. No SoCFPGA enables the Denali NAND, because support for that was completely broken and I was waiting until the Denali NAND patches make it in before sending that platform which will use it. Indeed, DM_MTD should be enabled in platform config, not in general.
Hi Marek, On Thu, Jan 30, 2020 at 7:03 AM Marek Vasut <marex at denx.de> wrote: > > On 1/29/20 6:08 PM, Simon Goldschmidt wrote: > > Am 29.01.2020 um 16:55 schrieb Masahiro Yamada: > >> UCLASS_MTD is a better fit for NAND drivers. > >> > >> Make NAND_DENALI_DT depend on DM_MTD, which is needed to compile > >> drivers/mtd/mtd-uclass.c > >> > >> Also, make ARCH_SOCFPGA and ARCH_UNIPHIER select DM_MTD because > >> they use this driver. > >> > >> Signed-off-by: Masahiro Yamada <yamada.masahiro at socionext.com> > >> --- > >> > >> Changes in v2: > >> - new patch > >> > >> arch/arm/Kconfig | 2 ++ > >> drivers/mtd/nand/raw/Kconfig | 2 +- > >> drivers/mtd/nand/raw/denali_dt.c | 4 ++-- > >> 3 files changed, 5 insertions(+), 3 deletions(-) > >> > >> diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig > >> index 1236315168..d1c58667c7 100644 > >> --- a/arch/arm/Kconfig > >> +++ b/arch/arm/Kconfig > >> @@ -935,6 +935,7 @@ config ARCH_SOCFPGA > >> select ARM64 if TARGET_SOCFPGA_STRATIX10 || TARGET_SOCFPGA_AGILEX > >> select CPU_V7A if TARGET_SOCFPGA_GEN5 || TARGET_SOCFPGA_ARRIA10 > >> select DM > >> + select DM_MTD > > > > Does any socfpga actually enable the denali driver? Do we need to > > default to this instead of enabling it in a defconfig? > > > > If we need it, could you please change that to 'imply'? Not all configs > > will need this. > > No SoCFPGA enables the Denali NAND, because support for that was > completely broken and I was waiting until the Denali NAND patches make > it in before sending that platform which will use it. > > Indeed, DM_MTD should be enabled in platform config, not in general. Right. None of ARCH_SOCFPGA boards enables the Denali, It is up to you whichever defconfig you add the CONFIG option to. I still prefer "select DM_MTD" for ARCH_UNIPHIER because all uniphier_*_defconfig enables the Denali driver. I will drop "select DM_MTD" from ARCH_SOCFPGA. Is this OK for you ?
On 1/30/20 6:34 AM, Masahiro Yamada wrote: > Hi Marek, Hi, > On Thu, Jan 30, 2020 at 7:03 AM Marek Vasut <marex at denx.de> wrote: >> >> On 1/29/20 6:08 PM, Simon Goldschmidt wrote: >>> Am 29.01.2020 um 16:55 schrieb Masahiro Yamada: >>>> UCLASS_MTD is a better fit for NAND drivers. >>>> >>>> Make NAND_DENALI_DT depend on DM_MTD, which is needed to compile >>>> drivers/mtd/mtd-uclass.c >>>> >>>> Also, make ARCH_SOCFPGA and ARCH_UNIPHIER select DM_MTD because >>>> they use this driver. >>>> >>>> Signed-off-by: Masahiro Yamada <yamada.masahiro at socionext.com> >>>> --- >>>> >>>> Changes in v2: >>>> - new patch >>>> >>>> arch/arm/Kconfig | 2 ++ >>>> drivers/mtd/nand/raw/Kconfig | 2 +- >>>> drivers/mtd/nand/raw/denali_dt.c | 4 ++-- >>>> 3 files changed, 5 insertions(+), 3 deletions(-) >>>> >>>> diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig >>>> index 1236315168..d1c58667c7 100644 >>>> --- a/arch/arm/Kconfig >>>> +++ b/arch/arm/Kconfig >>>> @@ -935,6 +935,7 @@ config ARCH_SOCFPGA >>>> select ARM64 if TARGET_SOCFPGA_STRATIX10 || TARGET_SOCFPGA_AGILEX >>>> select CPU_V7A if TARGET_SOCFPGA_GEN5 || TARGET_SOCFPGA_ARRIA10 >>>> select DM >>>> + select DM_MTD >>> >>> Does any socfpga actually enable the denali driver? Do we need to >>> default to this instead of enabling it in a defconfig? >>> >>> If we need it, could you please change that to 'imply'? Not all configs >>> will need this. >> >> No SoCFPGA enables the Denali NAND, because support for that was >> completely broken and I was waiting until the Denali NAND patches make >> it in before sending that platform which will use it. >> >> Indeed, DM_MTD should be enabled in platform config, not in general. > > > Right. None of ARCH_SOCFPGA boards enables the Denali, > It is up to you whichever defconfig you add the CONFIG option to. > > I still prefer "select DM_MTD" for ARCH_UNIPHIER > because all uniphier_*_defconfig enables the Denali driver. > > > > I will drop "select DM_MTD" from ARCH_SOCFPGA. > > Is this OK for you ? Yes, thanks.
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 1236315168..d1c58667c7 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -935,6 +935,7 @@ config ARCH_SOCFPGA select ARM64 if TARGET_SOCFPGA_STRATIX10 || TARGET_SOCFPGA_AGILEX select CPU_V7A if TARGET_SOCFPGA_GEN5 || TARGET_SOCFPGA_ARRIA10 select DM + select DM_MTD select DM_SERIAL select ENABLE_ARM_SOC_BOOT0_HOOK if TARGET_SOCFPGA_GEN5 || TARGET_SOCFPGA_ARRIA10 select OF_CONTROL @@ -1548,6 +1549,7 @@ config ARCH_UNIPHIER select DM_GPIO select DM_I2C select DM_MMC + select DM_MTD select DM_RESET select DM_SERIAL select DM_USB diff --git a/drivers/mtd/nand/raw/Kconfig b/drivers/mtd/nand/raw/Kconfig index 7814d84ba0..23201ca720 100644 --- a/drivers/mtd/nand/raw/Kconfig +++ b/drivers/mtd/nand/raw/Kconfig @@ -116,7 +116,7 @@ config NAND_DENALI config NAND_DENALI_DT bool "Support Denali NAND controller as a DT device" select NAND_DENALI - depends on OF_CONTROL && DM + depends on OF_CONTROL && DM_MTD help Enable the driver for NAND flash on platforms using a Denali NAND controller as a DT device. diff --git a/drivers/mtd/nand/raw/denali_dt.c b/drivers/mtd/nand/raw/denali_dt.c index 587e480faa..759ad40e51 100644 --- a/drivers/mtd/nand/raw/denali_dt.c +++ b/drivers/mtd/nand/raw/denali_dt.c @@ -160,7 +160,7 @@ static int denali_dt_probe(struct udevice *dev) U_BOOT_DRIVER(denali_nand_dt) = { .name = "denali-nand-dt", - .id = UCLASS_MISC, + .id = UCLASS_MTD, .of_match = denali_nand_dt_ids, .probe = denali_dt_probe, .priv_auto_alloc_size = sizeof(struct denali_nand_info), @@ -171,7 +171,7 @@ void board_nand_init(void) struct udevice *dev; int ret; - ret = uclass_get_device_by_driver(UCLASS_MISC, + ret = uclass_get_device_by_driver(UCLASS_MTD, DM_GET_DRIVER(denali_nand_dt), &dev); if (ret && ret != -ENODEV)
UCLASS_MTD is a better fit for NAND drivers. Make NAND_DENALI_DT depend on DM_MTD, which is needed to compile drivers/mtd/mtd-uclass.c Also, make ARCH_SOCFPGA and ARCH_UNIPHIER select DM_MTD because they use this driver. Signed-off-by: Masahiro Yamada <yamada.masahiro at socionext.com> --- Changes in v2: - new patch arch/arm/Kconfig | 2 ++ drivers/mtd/nand/raw/Kconfig | 2 +- drivers/mtd/nand/raw/denali_dt.c | 4 ++-- 3 files changed, 5 insertions(+), 3 deletions(-)