On Fri, Jul 28, 2017 at 11:08:58PM +0800, Bin Meng wrote:
Hi Tom,
On Fri, Jul 28, 2017 at 10:39 PM, Tom Rini trini@konsulko.com wrote:
Migrate all remaining instances of CMD_NAND, CMD_NAND_TRIMFFS CMD_NAND_LOCK_UNLOCK and CMD_NAND_TORTURE from the headers into the defconfig files.
Signed-off-by: Tom Rini trini@konsulko.com
[snip]
diff --git a/configs/MPC8569MDS_ATM_defconfig b/configs/MPC8569MDS_ATM_defconfig index b713db43adde..d4327c9817b9 100644 --- a/configs/MPC8569MDS_ATM_defconfig +++ b/configs/MPC8569MDS_ATM_defconfig @@ -10,6 +10,7 @@ CONFIG_BOOTDELAY=10 CONFIG_BOARD_EARLY_INIT_F=y CONFIG_HUSH_PARSER=y CONFIG_CMD_MMC=y +CONFIG_CMD_NAND=y
This is inconsistent. Why not imply it in arch/powerpc/cpu/mpc85xx/Kconfig::ARCH_MPC8569?
CONFIG_CMD_I2C=y CONFIG_CMD_DHCP=y CONFIG_CMD_MII=y diff --git a/configs/MPC8569MDS_defconfig b/configs/MPC8569MDS_defconfig index 96f7755f5de2..4f28bfe84698 100644 --- a/configs/MPC8569MDS_defconfig +++ b/configs/MPC8569MDS_defconfig @@ -9,6 +9,7 @@ CONFIG_BOOTDELAY=10 CONFIG_BOARD_EARLY_INIT_F=y CONFIG_HUSH_PARSER=y CONFIG_CMD_MMC=y +CONFIG_CMD_NAND=y CONFIG_CMD_I2C=y CONFIG_CMD_DHCP=y CONFIG_CMD_MII=y diff --git a/configs/P1020RDB-PC_36BIT_NAND_defconfig b/configs/P1020RDB-PC_36BIT_NAND_defconfig index 58723535b6c9..ed196af265ea 100644 --- a/configs/P1020RDB-PC_36BIT_NAND_defconfig +++ b/configs/P1020RDB-PC_36BIT_NAND_defconfig @@ -23,6 +23,7 @@ CONFIG_TPL_I2C_SUPPORT=y CONFIG_TPL_MPC8XXX_INIT_DDR_SUPPORT=y CONFIG_HUSH_PARSER=y CONFIG_CMD_MMC=y +CONFIG_CMD_NAND=y
arch/powerpc/cpu/mpc85xx/Kconfig::ARCH_P1020?
Anyway, if these are all handled by moveconfig, I don't quite understand how it handle these differently by adding some imply while updating some defconfigs?
I'm not 100% happy with the imply suggestions here. In both of these cases, moveconfig.py -i CMD_NAND -I min2 did suggest both of these, but longer term, and I should have mentioned in the commit message, once we have a 'NAND' option in Kconfig, CMD_NAND should be default y on that and ARCH_xxx should imply NAND when the SoC supports NAND and then yes, it's clear that ARCH_P1020, etc, should imply NAND.