[PATCH 13/13] ARM: Exynos: extend soft-reset support for Exynos5420
Chander Kashyap
chander.kashyap at linaro.org
Tue Jun 11 09:24:32 EDT 2013
On 10 June 2013 14:24, Kukjin Kim <kgene.kim at samsung.com> wrote:
> Tushar Behera wrote:
>>
>> On 06/06/2013 04:31 PM, Chander Kashyap wrote:
>> > Extend the soft reset support for Exynos5420 SoC.
>> >
>> > Signed-off-by: Chander Kashyap <chander.kashyap at linaro.org>
>> > ---
>> > arch/arm/mach-exynos/common.c | 3 ++-
>> > 1 file changed, 2 insertions(+), 1 deletion(-)
>> >
>> > diff --git a/arch/arm/mach-exynos/common.c b/arch/arm/mach-exynos/common.c
>> > index fbe1aa7..dc7c134 100644
>> > --- a/arch/arm/mach-exynos/common.c
>> > +++ b/arch/arm/mach-exynos/common.c
>> > @@ -321,7 +321,8 @@ void exynos5_restart(char mode, const char *cmd)
>> > u32 val;
>> > void __iomem *addr;
>> >
>> > - if (of_machine_is_compatible("samsung,exynos5250")) {
>> > + if (of_machine_is_compatible("samsung,exynos5250") ||
>> > + (of_machine_is_compatible("samsung,exynos5420"))) {
>> > val = 0x1;
>> > addr = EXYNOS_SWRESET;
>> > } else if (of_machine_is_compatible("samsung,exynos5440")) {
>> >
>>
>> How about making 0x1 and EXYNOS_SWRESET as default values of val and
>> addr. That way we can remove this check for exynos5250 and exynos5420.
>> Also we can remove the trailing else statement as there are no non-DT
>> boards for EXYNOS5.
>>
>> void exynos5_restart(char mode, const char *cmd)
>> {
>> struct device_node *np;
>> - u32 val;
>> - void __iomem *addr;
>> + u32 val = 0x1;
>> + void __iomem *addr = EXYNOS_SWRESET;
>>
>> - if (of_machine_is_compatible("samsung,exynos5250")) {
>> - val = 0x1;
>> - addr = EXYNOS_SWRESET;
>> - } else if (of_machine_is_compatible("samsung,exynos5440")) {
>> + if (of_machine_is_compatible("samsung,exynos5440")) {
>> u32 status;
>> np = of_find_compatible_node(NULL, NULL,
>> "samsung,exynos5440-clock");
>>
>> @@ -336,9 +333,6 @@ void exynos5_restart(char mode, const char *cmd)
>> val = __raw_readl(addr);
>>
>> val = (val & 0xffff0000) | (status & 0xffff);
>> - } else {
>> - pr_err("%s: cannot support non-DT\n", __func__);
>> - return;
>> }
>>
>> --
>
> Yeah, looks better.
Yes i will do the change. Thanks
>
> - Kukjin
>
--
with warm regards,
Chander Kashyap
More information about the linux-arm-kernel
mailing list