[RFC PATCH v2 02/12] of: Allow scripts/dtc/libfdt to be used from kernel code
Grant Likely
grant.likely at secretlab.ca
Sat Mar 5 11:57:50 EST 2011
[cc'ing David Gibson]
On Fri, Mar 04, 2011 at 11:42:14AM -0800, David Daney wrote:
> Signed-off-by: David Daney <ddaney at caviumnetworks.com>
> ---
> include/linux/libfdt.h | 3 +++
> lib/Kconfig | 6 ++++++
> lib/Makefile | 2 ++
> lib/libfdt/Makefile | 7 +++++++
> lib/libfdt/libfdt_env.h | 21 +++++++++++++++++++++
> scripts/dtc/libfdt/libfdt.h | 4 ++--
> 6 files changed, 41 insertions(+), 2 deletions(-)
> create mode 100644 include/linux/libfdt.h
> create mode 100644 lib/libfdt/Makefile
> create mode 100644 lib/libfdt/libfdt_env.h
>
> diff --git a/include/linux/libfdt.h b/include/linux/libfdt.h
> new file mode 100644
> index 0000000..10bce91
> --- /dev/null
> +++ b/include/linux/libfdt.h
> @@ -0,0 +1,3 @@
> +#include "../../lib/libfdt/libfdt_env.h"
libfdt_env.h should be in include/linux
> +#include "../../scripts/dtc/libfdt/fdt.h"
> +#include "../../scripts/dtc/libfdt/libfdt.h"
Hmmm... I wonder if there is a better way to do this. I don't much
care for the relative path references.
Also, need to have #ifdef _INCLUDE_LIBFDT_H_ protection
> diff --git a/lib/Kconfig b/lib/Kconfig
> index 0ee67e0..e8a2638 100644
> --- a/lib/Kconfig
> +++ b/lib/Kconfig
> @@ -219,4 +219,10 @@ config LRU_CACHE
> config AVERAGE
> bool
>
> +#
> +# The Flattened Device Tree manipulation library
> +#
> +config LIBFDT
> + bool
> +
This should be in drivers/of/Kconfig
> endmenu
> diff --git a/lib/Makefile b/lib/Makefile
> index cbb774f..5840115 100644
> --- a/lib/Makefile
> +++ b/lib/Makefile
> @@ -110,6 +110,8 @@ obj-$(CONFIG_ATOMIC64_SELFTEST) += atomic64_test.o
>
> obj-$(CONFIG_AVERAGE) += average.o
>
> +obj-$(CONFIG_LIBFDT) += libfdt/
> +
Ditto here; drivers/of/libfdt
> hostprogs-y := gen_crc32table
> clean-files := crc32table.h
>
> diff --git a/lib/libfdt/Makefile b/lib/libfdt/Makefile
> new file mode 100644
> index 0000000..6c1a496
> --- /dev/null
> +++ b/lib/libfdt/Makefile
> @@ -0,0 +1,7 @@
> +obj-y = fdt.o fdt_wip.o fdt_ro.o
> +
> +EXTRA_CFLAGS += -include $(src)/libfdt_env.h -I$(src)/../../scripts/dtc/libfdt
> +
> +$(obj)/%.o: $(src)/../../scripts/dtc/libfdt/%.c FORCE
> + $(call cmd,force_checksrc)
> + $(call if_changed_rule,cc_o_c)
> diff --git a/lib/libfdt/libfdt_env.h b/lib/libfdt/libfdt_env.h
> new file mode 100644
> index 0000000..d977b8b
> --- /dev/null
> +++ b/lib/libfdt/libfdt_env.h
> @@ -0,0 +1,21 @@
> +#ifndef _LIBFDT_ENV_H
> +#define _LIBFDT_ENV_H
> +
> +#include <linux/string.h>
> +
> +#define _B(n) ((unsigned long long)((uint8_t *)&x)[n])
> +static inline uint32_t fdt32_to_cpu(uint32_t x)
> +{
> + return (_B(0) << 24) | (_B(1) << 16) | (_B(2) << 8) | _B(3);
> +}
> +#define cpu_to_fdt32(x) fdt32_to_cpu(x)
> +
> +static inline uint64_t fdt64_to_cpu(uint64_t x)
> +{
> + return (_B(0) << 56) | (_B(1) << 48) | (_B(2) << 40) | (_B(3) << 32)
> + | (_B(4) << 24) | (_B(5) << 16) | (_B(6) << 8) | _B(7);
> +}
> +#define cpu_to_fdt64(x) fdt64_to_cpu(x)
> +#undef _B
> +
This isn't necessary, the kernel already has efficient architecture
macros for converting endianess.
#define fdt32_to_cpu(x) be32_to_cpu(x)
#define cpu_to_fdt32(x) cpu_to_be32(x)
#define fdt64_to_cpu(x) be64_to_cpu(x)
#define cpu_to_fdt64(x) cpu_to_be64(x)
> +#endif /* _LIBFDT_ENV_H */
> diff --git a/scripts/dtc/libfdt/libfdt.h b/scripts/dtc/libfdt/libfdt.h
> index ce80e4f..33a0c4d 100644
> --- a/scripts/dtc/libfdt/libfdt.h
> +++ b/scripts/dtc/libfdt/libfdt.h
> @@ -51,8 +51,8 @@
> * EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
> */
>
> -#include <libfdt_env.h>
> -#include <fdt.h>
> +#include "libfdt_env.h"
> +#include "fdt.h"
This causes problems. libfdt is an external library that is
periodically copied into the kernel tree. It would be better to add
"-Iscripts/dtc/libfdt" to CFLAGS for .c files that want to call libfdt
routines.
>
> #define FDT_FIRST_SUPPORTED_VERSION 0x10
> #define FDT_LAST_SUPPORTED_VERSION 0x11
> --
> 1.7.2.3
>
More information about the devicetree-discuss
mailing list