linux.git/drivers/of/fdt.c, branch v5.3

of/fdt: pass early_init_dt_reserve_memory_arch() with bool type nomap

2019-07-08T21:28:15+00:00

The third argument 'nomap' of early_init_dt_reserve_memory_arch() is
bool. It is preferred to pass it with a bool type parameter.

Signed-off-by: Masahiro Yamada 
Reviewed-by: Frank Rowand 
Signed-off-by: Rob Herring

of/fdt: Fix ‘of_fdt_match’ defined but not used compiler warning

2019-06-18T14:10:43+00:00

When CONFIG_OF_EARLY_FLATTREE is disabled, there is a compiler
warning,

drivers/of/fdt.c:129:19: warning: ‘of_fdt_match’ defined but not used [-Wunused-function]
 static int __init of_fdt_match(const void *blob, unsigned long node,

Since the only caller of of_fdt_match() is of_flat_dt_match(),
let's move the body of of_fdt_match() into of_flat_dt_match()
and eliminate of_fdt_match().

Meanwhile, move of_fdt_is_compatible() under CONFIG_OF_EARLY_FLATTREE,
as all callers are over there.

Fixes: 9b4d2b635bd0 ("of/fdt: Remove dead code and mark functions with __init")
Cc: Frank Rowand 
Signed-off-by: Kefeng Wang 
Reviewed-by: Stephen Boyd 
Signed-off-by: Rob Herring

of/fdt: Mark initial_boot_params as __ro_after_init

2019-05-24T21:39:27+00:00

The FDT pointer, i.e. initial_boot_params, shouldn't be changed after
init. It's only set by boot code and then the only user of the FDT is
the raw sysfs reading API. Mark this pointer with __ro_after_init so
that the pointer can't be changed after init.

Cc: Hsin-Yi Wang 
Signed-off-by: Stephen Boyd 
Signed-off-by: Rob Herring

of/fdt: Remove dead code and mark functions with __init

2019-05-24T21:39:14+00:00

Some functions in here are never called, and others are only called
during __init. Remove the dead code and some dead exports for functions
that don't exist (I'm looking at you of_fdt_get_string!). Mark some
functions with __init so we can throw them away after we boot up and
poke at the FDT blob too.

Cc: Hsin-Yi Wang 
Signed-off-by: Stephen Boyd 
Signed-off-by: Rob Herring

of: del redundant type conversion

2019-04-29T17:50:20+00:00

The type of variable l in early_init_dt_scan_chosen is
int, there is no need to convert to int.

Signed-off-by: xiaojiangfeng 
Reviewed-by: Frank Rowand 
Signed-off-by: Rob Herring

treewide: add checks for the return value of memblock_alloc*()

2019-03-12T17:04:02+00:00

Add check for the return value of memblock_alloc*() functions and call
panic() in case of error.  The panic message repeats the one used by
panicing memblock allocators with adjustment of parameters to include
only relevant ones.

The replacement was mostly automated with semantic patches like the one
below with manual massaging of format strings.

  @@
  expression ptr, size, align;
  @@
  ptr = memblock_alloc(size, align);
  + if (!ptr)
  + 	panic("%s: Failed to allocate %lu bytes align=0x%lx\n", __func__, size, align);

[anders.roxell@linaro.org: use '%pa' with 'phys_addr_t' type]
  Link: http://lkml.kernel.org/r/20190131161046.21886-1-anders.roxell@linaro.org
[rppt@linux.ibm.com: fix format strings for panics after memblock_alloc]
  Link: http://lkml.kernel.org/r/1548950940-15145-1-git-send-email-rppt@linux.ibm.com
[rppt@linux.ibm.com: don't panic if the allocation in sparse_buffer_init fails]
  Link: http://lkml.kernel.org/r/20190131074018.GD28876@rapoport-lnx
[akpm@linux-foundation.org: fix xtensa printk warning]
Link: http://lkml.kernel.org/r/1548057848-15136-20-git-send-email-rppt@linux.ibm.com
Signed-off-by: Mike Rapoport 
Signed-off-by: Anders Roxell 
Reviewed-by: Guo Ren 		[c-sky]
Acked-by: Paul Burton 		[MIPS]
Acked-by: Heiko Carstens 	[s390]
Reviewed-by: Juergen Gross 		[Xen]
Reviewed-by: Geert Uytterhoeven 	[m68k]
Acked-by: Max Filippov 		[xtensa]
Cc: Catalin Marinas 
Cc: Christophe Leroy 
Cc: Christoph Hellwig 
Cc: "David S. Miller" 
Cc: Dennis Zhou 
Cc: Greentime Hu 
Cc: Greg Kroah-Hartman 
Cc: Guan Xuetao 
Cc: Guo Ren 
Cc: Mark Salter 
Cc: Matt Turner 
Cc: Michael Ellerman 
Cc: Michal Simek 
Cc: Petr Mladek 
Cc: Richard Weinberger 
Cc: Rich Felker 
Cc: Rob Herring 
Cc: Rob Herring 
Cc: Russell King 
Cc: Stafford Horne 
Cc: Tony Luck 
Cc: Vineet Gupta 
Cc: Yoshinori Sato 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

of: Remove struct device_node.type pointer

2019-01-10T22:24:44+00:00

Now that all users of device_node.type pointer have been removed in
favor of accessor functions, we can remove it.

Cc: Frank Rowand 
Cc: devicetree@vger.kernel.org
Signed-off-by: Rob Herring

OF: Add a warning in case chosen node is not present

2018-12-03T23:41:59+00:00

On architectures that only get their bootargs through devicetree's
chosen node (such as RISC-V), that node is mandatory. After a
discussion with Rob [1] I'm adding a warning in case chosen node
is not present, to let users know about it.

[1]: https://patchwork.ozlabs.org/patch/984224/#2016136

Signed-off-by: Nick Kossifidis 
Reviewed-by: Palmer Dabbelt 
Signed-off-by: Rob Herring

of/fdt: Remove custom __early_init_dt_declare_initrd() implementation

2018-11-26T21:50:39+00:00

Now that ARM64 uses phys_initrd_start/phys_initrd_size, we can get rid
of its custom __early_init_dt_declare_initrd() which causes a fair
amount of objects rebuild when changing CONFIG_BLK_DEV_INITRD. In order
to make sure ARM64 does not produce a BUG() when VM debugging is turned
on though, we must avoid early calls to __va() which is what
__early_init_dt_declare_initrd() does and wrap this around to avoid
running that code on ARM64.

Signed-off-by: Florian Fainelli 
Reviewed-by: Mike Rapoport 
Signed-off-by: Rob Herring

of/fdt: Populate phys_initrd_start/phys_initrd_size from FDT

2018-11-26T21:50:26+00:00

Now that we have central and global variables holding the physical
address and size of the initrd, we can have
early_init_dt_check_for_initrd() populate
phys_initrd_start/phys_initrd_size for us.

This allows us to remove a chunk of code from arch/arm/mm/init.c
introduced with commit 65939301acdb ("arm: set initrd_start/initrd_end
for fdt scan").

Signed-off-by: Florian Fainelli 
Reviewed-by: Mike Rapoport 
Signed-off-by: Rob Herring