Message ID | 1439668968-3882-2-git-send-email-hdegoede@redhat.com |
---|---|
State | Accepted |
Delegated to: | Hans de Goede |
Headers | show |
On Sat, 2015-08-15 at 22:02 +0200, Hans de Goede wrote: > CONFIG_SPL_NAND_SUPPORT gets used via IS_ENABLED so it must be > defined > to 1, rather then just being defined. > > While at remove 2 other unused NAND related defines from sunxi > -common.h. > > Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
On Sat, Aug 15, 2015 at 10:02:34PM +0200, Hans de Goede wrote: > CONFIG_SPL_NAND_SUPPORT gets used via IS_ENABLED so it must be defined > to 1, rather then just being defined. > > While at remove 2 other unused NAND related defines from sunxi-common.h. > > Signed-off-by: Hans de Goede <hdegoede@redhat.com> > --- > include/configs/sunxi-common.h | 7 +------ > 1 file changed, 1 insertion(+), 6 deletions(-) > > diff --git a/include/configs/sunxi-common.h b/include/configs/sunxi-common.h > index 1abf73c..3735afb 100644 > --- a/include/configs/sunxi-common.h > +++ b/include/configs/sunxi-common.h > @@ -138,15 +138,10 @@ > #define CONFIG_SERIAL_TAG > > #if defined(CONFIG_SPL_NAND_SUNXI) > -#define CONFIG_SPL_NAND_DRIVERS > -#define CONFIG_SPL_NAND_SUPPORT > - > -#define CONFIG_SYS_NAND_SPL_KERNEL_OFFS 0x280000 > +#define CONFIG_SPL_NAND_SUPPORT 1 Every other platform just use an empty define for this one, and the only user of IS_ENABLED together with this option is the sunxi board. Why not simply replace the IS_ENABLED by an ifdef to maintain consistency? AFAIK, IS_ENABLED doesn't bring anythnig to the table here. Maxime
Hi, On 08/19/2015 03:45 PM, Maxime Ripard wrote: > On Sat, Aug 15, 2015 at 10:02:34PM +0200, Hans de Goede wrote: >> CONFIG_SPL_NAND_SUPPORT gets used via IS_ENABLED so it must be defined >> to 1, rather then just being defined. >> >> While at remove 2 other unused NAND related defines from sunxi-common.h. >> >> Signed-off-by: Hans de Goede <hdegoede@redhat.com> >> --- >> include/configs/sunxi-common.h | 7 +------ >> 1 file changed, 1 insertion(+), 6 deletions(-) >> >> diff --git a/include/configs/sunxi-common.h b/include/configs/sunxi-common.h >> index 1abf73c..3735afb 100644 >> --- a/include/configs/sunxi-common.h >> +++ b/include/configs/sunxi-common.h >> @@ -138,15 +138,10 @@ >> #define CONFIG_SERIAL_TAG >> >> #if defined(CONFIG_SPL_NAND_SUNXI) >> -#define CONFIG_SPL_NAND_DRIVERS >> -#define CONFIG_SPL_NAND_SUPPORT >> - >> -#define CONFIG_SYS_NAND_SPL_KERNEL_OFFS 0x280000 >> +#define CONFIG_SPL_NAND_SUPPORT 1 > > Every other platform just use an empty define for this one, and the > only user of IS_ENABLED together with this option is the sunxi > board. Why not simply replace the IS_ENABLED by an ifdef to maintain > consistency? Because u-boot is moving to Kconfig which will define it to 1, and ... > AFAIK, IS_ENABLED doesn't bring anythnig to the table here. IS_ENABLED leads too cleaner / easier to read code, so it does bring something to the table. REgards, Hans
On Thu, Aug 20, 2015 at 08:33:01AM +0200, Hans de Goede wrote: > Hi, > > On 08/19/2015 03:45 PM, Maxime Ripard wrote: > >On Sat, Aug 15, 2015 at 10:02:34PM +0200, Hans de Goede wrote: > >>CONFIG_SPL_NAND_SUPPORT gets used via IS_ENABLED so it must be defined > >>to 1, rather then just being defined. > >> > >>While at remove 2 other unused NAND related defines from sunxi-common.h. > >> > >>Signed-off-by: Hans de Goede <hdegoede@redhat.com> > >>--- > >> include/configs/sunxi-common.h | 7 +------ > >> 1 file changed, 1 insertion(+), 6 deletions(-) > >> > >>diff --git a/include/configs/sunxi-common.h b/include/configs/sunxi-common.h > >>index 1abf73c..3735afb 100644 > >>--- a/include/configs/sunxi-common.h > >>+++ b/include/configs/sunxi-common.h > >>@@ -138,15 +138,10 @@ > >> #define CONFIG_SERIAL_TAG > >> > >> #if defined(CONFIG_SPL_NAND_SUNXI) > >>-#define CONFIG_SPL_NAND_DRIVERS > >>-#define CONFIG_SPL_NAND_SUPPORT > >>- > >>-#define CONFIG_SYS_NAND_SPL_KERNEL_OFFS 0x280000 > >>+#define CONFIG_SPL_NAND_SUPPORT 1 > > > >Every other platform just use an empty define for this one, and the > >only user of IS_ENABLED together with this option is the sunxi > >board. Why not simply replace the IS_ENABLED by an ifdef to maintain > >consistency? > > Because u-boot is moving to Kconfig which will define it to 1, > and ... > > >AFAIK, IS_ENABLED doesn't bring anythnig to the table here. > > IS_ENABLED leads too cleaner / easier to read code, so it does > bring something to the table. Ok, if the plan is to move to Kconfig eventually, I guess it makes sense. Maxime
diff --git a/include/configs/sunxi-common.h b/include/configs/sunxi-common.h index 1abf73c..3735afb 100644 --- a/include/configs/sunxi-common.h +++ b/include/configs/sunxi-common.h @@ -138,15 +138,10 @@ #define CONFIG_SERIAL_TAG #if defined(CONFIG_SPL_NAND_SUNXI) -#define CONFIG_SPL_NAND_DRIVERS -#define CONFIG_SPL_NAND_SUPPORT - -#define CONFIG_SYS_NAND_SPL_KERNEL_OFFS 0x280000 +#define CONFIG_SPL_NAND_SUPPORT 1 #define CONFIG_SYS_NAND_U_BOOT_OFFS 0x008000 - #endif - /* mmc config */ #if !defined(CONFIG_UART0_PORT_F) #define CONFIG_MMC
CONFIG_SPL_NAND_SUPPORT gets used via IS_ENABLED so it must be defined to 1, rather then just being defined. While at remove 2 other unused NAND related defines from sunxi-common.h. Signed-off-by: Hans de Goede <hdegoede@redhat.com> --- include/configs/sunxi-common.h | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-)