| Message ID | 1409680587-29818-7-git-send-email-will.deacon@arm.com |
|---|---|
| State | New |
| Headers | show |
On Tuesday 02 September 2014 18:56:26 Will Deacon wrote: > @@ -803,9 +804,11 @@ static int __init customize_machine(void) > if (machine_desc->init_machine) > machine_desc->init_machine(); > #ifdef CONFIG_OF > - else > + else { > + of_iommu_init(); > of_platform_populate(NULL, of_default_bus_match_table, > NULL, NULL); > + } > #endif > return 0; > } I think it would be better to not have this depend on the presence of an init_machine callback. Why not move of_iommu_init() in front of the if()? It shouldn't do anything unless we have an iommu driver that we should be using. Arnd
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 84db893dedc2..38a0203523af 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -18,6 +18,7 @@ #include <linux/bootmem.h> #include <linux/seq_file.h> #include <linux/screen_info.h> +#include <linux/of_iommu.h> #include <linux/of_platform.h> #include <linux/init.h> #include <linux/kexec.h> @@ -803,9 +804,11 @@ static int __init customize_machine(void) if (machine_desc->init_machine) machine_desc->init_machine(); #ifdef CONFIG_OF - else + else { + of_iommu_init(); of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL); + } #endif return 0; }
We need to ensure that the IOMMUs in the system have a chance to perform some basic initialisation before we start adding masters to them. This patch adds a call to of_iommu_init before of_platform_populate. Signed-off-by: Will Deacon <will.deacon@arm.com> --- arch/arm/kernel/setup.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)