Currently the main recursive tree printing function,
write_tree_source_node(), calls guess_type() to apply heuristics to
see how to print a property value, then calls the appropriate
write_propval_*() function to print it.
However, future heuristics for handling internal labels and the like
don't work well this way. Therefore, this patch refactors things to
have write_tree_source_node() call a new write_propval() function,
which incorporates the heurstic logic from guess_type() and also calls
the right function to do the actual printing.
No behavioural change.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
There is a bug in the sw_tree1 testcase / utility which puts two
"compatible" properties in one node in the output tree. This patch
fixes the bug, and also adds a new test checking that the sw_tree1
output is equal to test_tree1.dtb as its supposed to be, which should
catch future errors of this type.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch documents a few more functions in libfdt.h. It also makes
a slight update to the description of the FDT_ERR_INTERNAL error code.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
libfdt: Add more documentation (patch the third)
This patch adds documentation in libfdt.h for a few more libfdt
functions. It also makes a slight update to the documentation of
fdt_get_name().
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
/usr/bin/install: cannot stat `fdt.h': No such file or directory
/usr/bin/install: cannot stat `libfdt.h': No such file or directory
Signed-off-by: Emil Medve <Emilian.Medve@Freescale.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
At present, defining a property as, say:
foo = [abcd], <ffffffff>;
Will cause dtc to insert 2 bytes of zeros between the abcd and the
ffffffff, to align the cell form data.
Doing so seemed like a good idea at the time, but I don't believe
there are any users who actually rely on this behaviour. Segher
claims that OF has some defined bindings which include properties an
unaligned subsection of which is interpreted as 32-bit ints (i.e. like
cell data).
Worse, this alignment will cause nothing but pain when we add
expression support to dtc (when celldata is included in a larger
bytestring expession, we won't know the size of the preceding chunk of
the expression until it's evaluated, so we would have to carry
alignment fixup information right through the expression evaluation
process).
Therefore, this patch kills off this alignment behaviour.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch changes -Odts mode output so that labels on properties,
nodes and memreserve entries in input source are preserved in the
output.
Preserving labels within property values is trickier - another patch
coming later.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Remove the uneccessary LDFLAGS from the top-level makefile. It only
added libfdt/ to the link path. dtc doesn't need libfdt at all, and
the testcases which do, already link libfdt.a by explicit path.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch moves the dtc code for checking the device tree its
processing into a new checks.c. The tree accessor functions from
livetree.c which the checks use are exported and added to dtc.h.
Another small step towards a flexible checking architecture.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Currently all the read/write functions in libfdt require that the
given tree be v17, and further, that the tree has the memory
reservation block, structure block and strings block stored in that
physical order.
This patch eases these constraints, by making fdt_open_int() reorder
the blocks, and/or convert the tree to v17, so that it will then be
ready for the other read-write functions.
It also extends fdt_pack() to actually remove any gaps between blocks
that might be present.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
If future dtb version > 17 are defined, that are still backwards
compatible with v16, libfdt will of course be able to read and
understand them. However, when modifying such a tree, it can't
guarantee that it won't clobber additional structure from the new
version which it doesn't know about. Therefore, before making
modifications to a tree of version >17, we must change it's version to
be exactly 17.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Previously, there were a few shift/reduce and reduce/reduce
errors in the grammar that were being handled by the not-so-popular
GLR Parser technique.
Flip a right-recursive stack-abusing rule into a left-recursive
stack-friendly rule and clear up three messes in one shot: No more
conflicts, no need for the GLR parser, and friendlier stackness.
Compensate by reversing the property list on the node.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
At present, all the example dtbs we use in the testsuite are version
17 and have reservation map, then structure block then strings block
(the natural ordering based on alignment constraints). However, all
libfdt's read-only and in-place write functions should also work on
v16 trees, and on trees with other layouts.
This patch adds a testcase / utility function to rearrange the blocks
of a dtb and/or regress a v17 tree to v16, and uses it to run tests on
trees with different layouts and versions.
Signed-off-by: David Gibson <david@tgibson.dropbear.id.au>
The fdt_set_header() macro casts an arbitrary pointer into (struct
fdt_header *) to set fdt header fields. While we need to change the
type, so that we can use this macro on the usual (void *) used to
represent a device tree blob, the current macro also casts away any
const on the input pointer, which loses an important check.
This patch replaces the fdt_set_header() macro with a set of inline
functions, one for each header field which do a similar thing, but
which won't silently remove const from a given pointer. This approach
is also more in keeping with the individual accessor macros we use for
reading fdt header fields.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch adds some internal documentation in libfdt.h, in the form
of comments on the error codes and some functions. Only a couple of
functions are covered so far, leaving the documentation still woefully
inadequate, but hey, it's a start.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Although it's a low-level function that shouldn't normally be needed,
there are circumstances where it's useful for users of libfdt to use
the _fdt_next_tag() function. Therefore, this patch renames it to
fdt_next_tag() and publishes it in libfdt.h.
In addition, this patch adds a new testcase using fdt_next_tag(),
dtbs_equal_ordered. This testcase tests for structural equality of
two dtbs, including the order of properties and subnodes, but ignoring
NOP tags, the order of the dtb sections and the layout of strings in
the strings block. This will be useful for testing other dtc
functionality in the future.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
It's potentially useful for users of libfdt to sanity check a device
tree (or, rather, a blob of data which may or may not be a device
tree) before processing it in more detail with libfdt.
This patch renames the libfdt internal function _fdt_check_header() to
fdt_check_header() and makes it a published function, so it can now be
used for this purpose.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
I like to see the basis cases established early in
the rule sets, so place "empty" reduction first.
Purely cosmetic.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
At present, dtc makes a lot of semantic checks on the device tree by
default, and will refuse to produce output if they fail. This means
people tend to need -f to force output despite failing semantic checks
rather a lot.
This patch splits the device tree checks into structural checks (no
bad or duplicate names or phandles) and semantic checks (everything
else). By default, only the structural checks are performed, and are
fatal. -f will force output even with structural errors (using this
in -Idts mode would essentially always be a bad idea, but it might be
useful in -Idtb mode for examining a malformed dtb).
Semantic checks are only performed if the new -c command line option
is supplied, and are always warnings only. Semantic checks will never
be performed on a tree with structural errors.
This patch is only a stopgap before implementing proper fine-grained
error/warning handling, but it should at least get rid of the
far-too-frequent need for -f for the time being.
This patch removes the -f from the dtc testcases now that it's no
longer necessary.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
fdt_del_node(), unlike most of the rw functions does not check the
fdt's header with RW_CHECK_HEADER. However, it could make a mess of
things if the conditions in RW_CHECK_HEADER aren't met. So, this
patch adds the omitted check.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch makes the helper macros in trees.S use separate labels for
the end of each dt subblock, rather than using only start labels.
This means that the macros can now be used to create trees with the
subblocks in non-standard orders.
In addition, it adds a bunch of extra ; after lines of asm code in
macros, making them safe to use in nested macros.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Currently, fdt_string() returns a (non-const) char *, despite taking a
const void *fdt. This is inconsistent with all the other read-only
functions which all return const pointers into the blob.
This patch fixes that.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch makes improvements to the way properties are printed when
in dtc is producing dts output.
- Characters which need escaping are now properly handled when
printing properties as strings
- The heuristics for what format to use for a property are
improved so that 'compatible' properties will be displayed as
expected.
- escapes.dts is altered to better demonstrate the changes,
and the string_escapes testcase is adjusted accordingly.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Recent commits 333542fabf and
fd1bf3a5ae added new testcases to dtc.
However, although the testcases were added to the Makefile and
run_tests.sh, one of the .c files for the testcase was omitted from
the patch in each case.
This patch restores the missing testcase code.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
dtc supports the use of C-style escapes (\n, \t and so forth) in
string property definitions via the data_copy_escape_string()
function. However, while it supports the most common escape
characters, it doesn't support the full set that C does, which is a
potential gotcha.
Worse, a bug in the lexer means that while data_copy_escape_string()
can handle the \" escape, a string with such an escape won't lex
correctly.
This patch fixes both problems, extending data_copy_escape_string() to
support the missing escapes, and fixing the regex for strings in the
lexer to handle internal escaped quotes.
This also adds a testcase for string escape functionality.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch adds functions for dealing with the compatible property.
fdt_node_check_compatible() can be used to determine whether a node is
compatible with a given string and fdt_node_offset_by_compatible()
locates nodes with a given compatible string.
Testcases for these functions are also included.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The dtc/libfdt testsuite creates a number of .dtb files during its
run. To ensure a clean test run, these are currently deleted before
each group of tests.
This is, in fact, a mistake, since if something goes wrong in the
first group of tests, deleting the .dtb at the beginning of the second
group of tests makes it harder to figure out what the problem was.
This patch changes the script to only delete the files once, before
the whole test run.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
libfdt.h currently includes fdt.h, then libfdt_env.h. This is
incorrect, because depending on the environment into which libfdt is
embedded, libfdt_env.h may be needed to define datatypes used in
fdt.h. This patch corrects the problem.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch adds functions to libfdt for accessing the memory
reservation map section of a device tree blob. fdt_num_mem_rsv()
retreives the number of reservation entries in a dtb, and
fdt_get_mem_rsv() retreives a specific reservation entry.
fdt_add_mem_rsv() adds a new entry, and fdt_del_mem_rsv() removes a
specific numbered entry.
Testcases for these new functions are also included.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch makes a number of Makefile cleanups and improvements:
- We use more generic rules to invoke flex and bison, which is
useful for some of the other changes.
- We use the name dtc-lexer.lex.c for the flex output, instead
of the default lex.yy.c. That means less potential for confusion if
dtc is embedded into other projects (e.g. the kernel).
- We separate out a Makefile.dtc designed for embedding into
other projects, analagous to Makefile.libfdt.
- Makefile.libfdt is cleaned up to be more useful based on
some actual trial runs of embedding libfdt in the kernel bootwrapper.
- Versioning related rules and variables are collected into
one place in the Makefile.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
At present, the fdt_subnode_offset() and fdt_path_offset() functions
in libfdt require the exact name of the nodes in question be passed,
including unit address.
This is contrary to traditional OF-like finddevice() behaviour, which
allows the unit address to be omitted (which is useful when the device
name is unambiguous without the address).
This patch introduces similar behaviour to
fdt_subnode_offset_namelen(), and hence to fdt_subnode_offset() and
fdt_path_offset() which are implemented in terms of the former. The
unit address can be omitted from the given node name. If this is
ambiguous, the first such node in the flattened tree will be selected
(this behaviour is consistent with IEEE1275 which specifies only that
an arbitrary node matching the given information be selected).
This very small change is then followed by many more diffs which
change the test examples and testcases to exercise this behaviour.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
At present, the testcases for read/write functions (setprop,
del_property and del_node) are only invoked on the single
asm-generated tree, not on any of the other tree images which should
be equivalent. The functions in question will (correctly) not work on
the "unfinished" tree output from sw_tree1, but should work on most of
the others.
This patch extends the run_tests script to invoke the r/w testcases on
more example trees. The testsuite still passes clean with this
addition.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
In the dtc tree, both flat_dt.h and libfdt/fdt.h have structures and
constants relating to the flattened device tree format derived from
asm-powerpc/prom.h in the kernel. The former is used in dtc, the
latter in libfdt.
libfdt/fdt.h is the more recent, revised version, so use that
throughout, removing flat_dt.h.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This large patch removes all trailing whitespace from dtc (including
libfdt, the testsuite and documentation). It also removes a handful
of redundant blank lines (at the end of functions, or when there are
two blank lines together for no particular reason).
As well as anything else, this means that quilt won't whinge when I go
to convert the whole of libfdt into a patch to apply to the kernel.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch adds a handful of simple testcases for dtc. It adds a dts
file which should generate the same sample tree as is used for the
libfdt testcases, and tests invoking dtc on this dts, plus the
standard batch of libfdt cases on the resulting dtb, which effectively
checks that the dtb is correct.
Because the test framework assumes each testcase is an executable with
the right output conventions, we use a little shell script, dtc.sh, as
a wrapper around dtc itself. It simply invokes dtc and returns a PASS
or FAIL depending on whether dtc returned an error.
It's not much, but it's a start.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
run_tests.sh from the dtc testsuite currently has a facility ro run
just "functional" or just "stress" tests. This distinction is carried
over from libhugetlbfs where the test framework originated, and where
it made sense.
In dtc, we have no stress tests, so running these subsections isn't
particularly interesting. This patch removes these test subsets,
instead defining a single "libfdt" test subset for testcases related
to libfdt (and not dtc proper only. Currently that's all of the
testcases, but with any luck we'll have some dtc testcases in the
future.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Flat device trees always have integers in their structure stored as
big-endian. From this point of view, property values are
bags-of-bytes and any endianness is up to users of the device tree to
determine.
The libfdt testcases which use properties with integer values,
currently use native endian format for the architecture on which the
testcases are run. This works ok for now, since both the creation and
checking of the example device trees happen in the same endianness.
This will become a problem, however, for tests of dtc which we want to
add in the nearish future. dtc always uses big-endian format for
'cell' format data in properties; as it needs to in order to produce
powerpc-usable device trees when hosted on a little-endian
architecture.
This patch, therefore, changes the libfdt testsuite to use big-endian
format always for integer format data, in order to interoperate sanely
with future dtc testcases. This also means that the example trees
created by the testsuite should now be byte-for-byte identical
regardless of dtc and libfdt's host platform, which is arguably an
advantage.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch adds a function to libfdt to locate nodes containing a
property with a specific value.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
With kernel commit eff2ebd207af9f501af0ef667a7d14befcb36c1b, we
clarified that in the flattened tree format, a particular nodes
properties are required to precede its subdnodes.
At present however, both dtc and libfdt will process trees which don't
meet this condition. This patch simplifies the code for
fdt_get_property() based on assuming that constraint. dtc continues
to be able to handle such an invalid tree - on the grounds that it's
useful for dtc to be able to correct such a broken tree - but this
patch adds a warning when this condition is not met while reading a
flattened tree.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch turns on optimisation in the Makefile by default. With the
optimizer on, some uninitialized variable warnings (one real, two
bogus) are now generated. This patch also squashes those again.
Currently, dtc relies on make's implicit rule to build the testcases.
This means that when not making verbosely (V=0, the default) there is
no message at all while relinking the testsuites. This can be very
confusing when updating libfdt.a (upon which the testcases depend) and
make appears to do nothing.
This patch corrects the situation, borrowing the rule used to link dtc
itself to link all the testcases as well.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
My recent implemenetation of fdt_get_path() had a bug - the while loop
tested offset which was unitialized on the first iteration. Depending
on code surrounding the call, this could cause fdt_get_path() to
return incorrect results.
This patch corrects the problem by applying some more correct thinking
to the loop condition.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
David Gibson