linux.git/include/linux/of_fdt.h, branch v3.16-rc7

of/fdt: add FDT address translation support

2014-05-20T20:19:25+00:00

Copy u-boot's FDT address translation code from common/fdt_support. This
code was originally based on the kernel's unflattened DT address parsing
code.

This commit can be reverted once relicensing of this code to GPLv2/BSD
is done and it is added to libfdt.

Signed-off-by: Rob Herring 
Acked-by: Grant Likely

of: push struct boot_param_header and defines into powerpc

2014-04-30T05:59:20+00:00

Now powerpc is the only user of struct boot_param_header and FDT defines,
so they can be moved into the powerpc architecture code.

Signed-off-by: Rob Herring 
Acked-by: Benjamin Herrenschmidt 
Cc: Paul Mackerras 
Cc: linuxppc-dev@lists.ozlabs.org
Tested-by: Grant Likely 
Tested-by: Stephen Chivers

of/fdt: convert initial_boot_params to opaque pointer

2014-04-30T05:59:19+00:00

Now that all accesses to FDT header data has been converted to accessor
helpers, initial_boot_params can become an opaque pointer.

Signed-off-by: Rob Herring 
Tested-by: Michal Simek 
Tested-by: Grant Likely 
Tested-by: Stephen Chivers

of/fdt: introduce of_get_flat_dt_size

2014-04-30T05:59:18+00:00

Add a wrapper function to retrieve the FDT size from the FDT header. This
is primarily to avoid libfdt include paths for the whole kernel.

Signed-off-by: Rob Herring 
Tested-by: Grant Likely 
Tested-by: Stephen Chivers

of/fdt: use libfdt accessors for header data

2014-04-30T05:59:16+00:00

With libfdt support, we can take advantage of helper accessors in libfdt
for accessing the FDT header data. This makes the code more readable and
makes the FDT blob structure more opaque to the kernel. This also
prepares for removing struct boot_param_header completely.

Signed-off-by: Rob Herring 
Cc: Max Filippov 
Tested-by: Michal Simek 
Tested-by: Grant Likely 
Tested-by: Stephen Chivers

of/fdt: Convert FDT functions to use libfdt

2014-04-30T05:59:15+00:00

The kernel FDT functions predate libfdt and are much more limited in
functionality. Also, the kernel functions and libfdt functions are
not compatible with each other because they have different definitions
of node offsets. To avoid this incompatibility and in preparation to
add more FDT parsing functions which will need libfdt, let's first
convert the existing code to use libfdt.

The FDT unflattening, top-level FDT scanning, and property retrieval
functions are converted to use libfdt. The scanning code should be
re-worked to be more efficient and understandable by using libfdt to
find nodes directly by path or compatible strings.

Signed-off-by: Rob Herring 
Tested-by: Michal Simek 
Tested-by: Grant Likely 
Tested-by: Stephen Chivers

of/fdt: update of_get_flat_dt_prop in prep for libfdt

2014-04-30T05:59:15+00:00

Make of_get_flat_dt_prop arguments compatible with libfdt fdt_getprop
call in preparation to convert FDT code to use libfdt. Make the return
value const and the property length ptr type an int.

Signed-off-by: Rob Herring 
Tested-by: Michal Simek 
Tested-by: Grant Likely 
Tested-by: Stephen Chivers

of/fdt: remove unused of_scan_flat_dt_by_path

2014-04-30T05:59:14+00:00

of_scan_flat_dt_by_path is unused anywhere in the kernel, so remove it.

Signed-off-by: Rob Herring 
Tested-by: Michal Simek 
Tested-by: Grant Likely 
Tested-by: Stephen Chivers

of/fdt: consolidate built-in dtb section variables

2014-04-30T05:59:13+00:00

Unify the various architectures __dtb_start and __dtb_end definitions
moving them into of_fdt.h.

Signed-off-by: Rob Herring 
Acked-by: Vineet Gupta 
Acked-by: James Hogan 
Tested-by: Michal Simek 
Cc: Ralf Baechle 
Cc: Jonas Bonn 
Cc: Chris Zankel 
Cc: Max Filippov 
Cc: linux-metag@vger.kernel.org
Cc: linux-mips@linux-mips.org
Cc: linux@lists.openrisc.net
Cc: linux-xtensa@linux-xtensa.org
Tested-by: Grant Likely 
Tested-by: Stephen Chivers

drivers: of: add initialization code for static reserved memory

2014-03-11T17:26:47+00:00

This patch adds support for static (defined by 'reg' property) reserved
memory regions declared in device tree.

Memory blocks can be reliably reserved only during early boot. This must
happen before the whole memory management subsystem is initialized,
because we need to ensure that the given contiguous blocks are not yet
allocated by kernel. Also it must happen before kernel mappings for the
whole low memory are created, to ensure that there will be no mappings
(for reserved blocks). Typically, all this happens before device tree
structures are unflattened, so we need to get reserved memory layout
directly from fdt.

Based on previous code provided by Josh Cartwright 

Signed-off-by: Marek Szyprowski 
Signed-off-by: Grant Likely