* [RESEND PATCH v3 0/3] Introduce fdtgrep for subsetting and hashing FDTs
@ 2014-03-22 15:07 Simon Glass
[not found] ` <1395500836-12735-1-git-send-email-sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
0 siblings, 1 reply; 9+ messages in thread
From: Simon Glass @ 2014-03-22 15:07 UTC (permalink / raw)
To: Devicetree Compiler; +Cc: Jon Loeliger, David Gibson, Simon Glass
This series adds two new functions, fdt_first_region() and
fdt_next_regions() which map FDT parts such as nodes and properties to
their regions in the FDT binary. The function is then used to implement
a grep utility for FDTs.
The core code is quite simple and small, but it grew a little due
to the need to make it iterative (first/next/next). Also this series adds
tests and a grep utility, so quite a bit of code is built on it.
The use for this feature is twofold. Firstly it provides a convenient
way of performing a structure-aware grep of the tree. For example it is
possible to grep for a node and get all the properties associated with
that node. Trees can be subsetted easily, by specifying the nodes that
are required, and then writing out the regions returned by this function.
This is useful for small resource-constrained systems, such as boot
loaders, which want to use an FDT but do not need to know about all of
it. The full FDT can be grepped to pull out the few things that are
needed - this can be an automatic part of the build system and does
not require the FDT source code.
This first use makes it easy to implement an FDT grep. Options are
provided to search for matching nodes (by name or compatible string),
properties and also for any of the above. It is possible to search for
non-matches also (useful for excluding a particular property from the
FDT, for example). The output is like fdtdump, but only with the regions
selected by the grep. Options are also provided to print the string table,
memory reservation table, etc. The fdtgrep utility can output valid
source, which can be used by dtc, but it can also directly output a new
.dtb binary.
Secondly it makes it easy to hash parts of the tree and detect changes.
The intent is to get a list of regions which will be invariant provided
those parts are invariant. For example, if you request a list of regions
for all nodes but exclude the property "data", then you will get the
same region contents regardless of any change to "data" properties.
An assumption is made here that the tree ordering remains the same.
This second use is the subject of a recent series sent to the U-Boot
mailing list, to enhance FIT images to support verified boot. Briefly,
this works by signing configurations (consisting of particular kernel
and FDT combinations) so that the boot loader can verify that these
combinations are valid and permitted. Since a FIT image is in fact an
FDT, we need to be able to hash particular regions of the FDT for the
signing and verification process. This is done by using the region functions
to select the data that needs to be hashed for a particular configuration.
The fdtgrep utility could be used to replace all of the functions of
fdtdump. However fdtdump is intended as a separate, simple way of
dumping the tree (for verifying against dtc output for example). So
fdtdump remains a separate program and this series leaves it alone.
Note: a somewhat unfortunately feature of this implementation is that
a state structure needs to be kept around between calls of
fdt_next_region(). This is declared in libfdt.h but really should be
opaque.
Changes in v3:
- Add a feature to include all subnodes
- Adjust help and command line processing to follow new approach
- Rename -V to -I to avoid using -V for a different purpose to other tools
- Rename -s to -e since it only 'enters' the node and does not include it all
- Add -s option to include all subnodes
- Add -f option to display offset; make -a display an absolute file address
- Add documentation on fdtgrep
Changes in v2:
- Move region code to separate fdt_region.c file
- Fix info->count <= info->max_regions in fdt_add_region() merge case
- Add new FDT_ERR_TOODEEP error type and use it
- Change returned error from BADLAYOUT to BADSTRUCTURE
- Return FDT_ERR_BADLAYOUT error if strings block is before structure block
- Add note that changes in node/property order can cause false hash misses
- Add more comments about the -1 return value from h_include
- Drop FDT_IS_COMPAT and pass node offset to h_include function
- Drop stale comment about names / wildcards
- Move to a model with fdt_first_region()/fdt_next_region()
- Add long comment explaining theory of operation
- Add local fdt_find_regions() function since libfdt no longer has it
Simon Glass (3):
libfdt: Add function to find regions in an FDT
Add documentation for fdtget/put
Add fdtgrep to grep and subset FDTs
.gitignore | 1 +
Documentation/manual.txt | 125 +++++++
Makefile | 4 +
Makefile.utils | 7 +
fdtgrep.c | 873 +++++++++++++++++++++++++++++++++++++++++++++++
libfdt/Makefile.libfdt | 3 +-
libfdt/fdt_region.c | 457 +++++++++++++++++++++++++
libfdt/libfdt.h | 217 +++++++++++-
tests/.gitignore | 1 +
tests/Makefile.tests | 3 +-
tests/grep.dts | 23 ++
tests/region_tree.c | 352 +++++++++++++++++++
tests/run_tests.sh | 361 +++++++++++++++++++-
tests/tests.sh | 1 +
14 files changed, 2424 insertions(+), 4 deletions(-)
create mode 100644 fdtgrep.c
create mode 100644 libfdt/fdt_region.c
create mode 100644 tests/grep.dts
create mode 100644 tests/region_tree.c
--
1.9.1.423.g4596e3a
--
To unsubscribe from this list: send the line "unsubscribe devicetree-compiler" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 9+ messages in thread
* [RESEND PATCH v3 1/3] libfdt: Add function to find regions in an FDT
[not found] ` <1395500836-12735-1-git-send-email-sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
@ 2014-03-22 15:07 ` Simon Glass
[not found] ` <1395500836-12735-2-git-send-email-sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
2014-03-22 15:07 ` [RESEND PATCH v3 2/3] Add documentation for fdtget/put Simon Glass
2014-03-22 15:07 ` [RESEND PATCH v3 3/3] Add fdtgrep to grep and subset FDTs Simon Glass
2 siblings, 1 reply; 9+ messages in thread
From: Simon Glass @ 2014-03-22 15:07 UTC (permalink / raw)
To: Devicetree Compiler; +Cc: Jon Loeliger, David Gibson, Simon Glass
Given a set of nodes and properties, find the regions of the device tree
which describe those parts.
A test is provided which builds a tree while tracking where the regions
should be, then calls fdt_first/next_region() to make sure that it agrees.
Further tests will come as part of fdtgrep.
Signed-off-by: Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
---
Changes in v3:
- Add a feature to include all subnodes
Changes in v2:
- Move region code to separate fdt_region.c file
- Fix info->count <= info->max_regions in fdt_add_region() merge case
- Add new FDT_ERR_TOODEEP error type and use it
- Change returned error from BADLAYOUT to BADSTRUCTURE
- Return FDT_ERR_BADLAYOUT error if strings block is before structure block
- Add note that changes in node/property order can cause false hash misses
- Add more comments about the -1 return value from h_include
- Drop FDT_IS_COMPAT and pass node offset to h_include function
- Drop stale comment about names / wildcards
- Move to a model with fdt_first_region()/fdt_next_region()
- Add long comment explaining theory of operation
libfdt/Makefile.libfdt | 3 +-
libfdt/fdt_region.c | 457 +++++++++++++++++++++++++++++++++++++++++++++++++
libfdt/libfdt.h | 217 ++++++++++++++++++++++-
tests/.gitignore | 1 +
tests/Makefile.tests | 3 +-
tests/region_tree.c | 352 +++++++++++++++++++++++++++++++++++++
tests/run_tests.sh | 5 +
7 files changed, 1035 insertions(+), 3 deletions(-)
create mode 100644 libfdt/fdt_region.c
create mode 100644 tests/region_tree.c
diff --git a/libfdt/Makefile.libfdt b/libfdt/Makefile.libfdt
index 91126c0..84769a4 100644
--- a/libfdt/Makefile.libfdt
+++ b/libfdt/Makefile.libfdt
@@ -6,5 +6,6 @@
LIBFDT_soname = libfdt.$(SHAREDLIB_EXT).1
LIBFDT_INCLUDES = fdt.h libfdt.h libfdt_env.h
LIBFDT_VERSION = version.lds
-LIBFDT_SRCS = fdt.c fdt_ro.c fdt_wip.c fdt_sw.c fdt_rw.c fdt_strerror.c fdt_empty_tree.c
+LIBFDT_SRCS = fdt.c fdt_ro.c fdt_wip.c fdt_sw.c fdt_region.c
+LIBFDT_SRCS += fdt_rw.c fdt_strerror.c fdt_empty_tree.c
LIBFDT_OBJS = $(LIBFDT_SRCS:%.c=%.o)
diff --git a/libfdt/fdt_region.c b/libfdt/fdt_region.c
new file mode 100644
index 0000000..3bb550d
--- /dev/null
+++ b/libfdt/fdt_region.c
@@ -0,0 +1,457 @@
+/*
+ * libfdt - Flat Device Tree manipulation
+ * Copyright (C) 2013 Google, Inc
+ *
+ * libfdt is dual licensed: you can use it either under the terms of
+ * the GPL, or the BSD license, at your option.
+ *
+ * a) This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this library; if not, write to the Free
+ * Software Foundation, Inc., 51 Franklin St, Fifth Floor, Boston,
+ * MA 02110-1301 USA
+ *
+ * Alternatively,
+ *
+ * b) Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ * 2. Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
+ * CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
+ * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+ * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
+ * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+ * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+ * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
+ * EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include "libfdt_env.h"
+
+#include <fdt.h>
+#include <libfdt.h>
+
+#include "libfdt_internal.h"
+
+/**
+ * fdt_add_region() - Add a new region to our list
+ *
+ * The region is added if there is space, but in any case we increment the
+ * count. If permitted, and the new region overlaps the last one, we merge
+ * them.
+ *
+ * @info: State information
+ * @offset: Start offset of region
+ * @size: Size of region
+ */
+static int fdt_add_region(struct fdt_region_state *info, int offset, int size)
+{
+ struct fdt_region *reg = &info->region[info->count - 1];
+
+ if (info->can_merge && info->count &&
+ info->count <= info->max_regions &&
+ offset <= reg->offset + reg->size) {
+ reg->size = offset + size - reg->offset;
+ } else if (info->count++ < info->max_regions) {
+ reg++;
+ reg->offset = offset;
+ reg->size = size;
+ } else {
+ return -1;
+ }
+
+ return 0;
+}
+
+/**
+ * fdt_include_supernodes() - Include supernodes required by this node
+ *
+ * When we decided to include a node or property which is not at the top
+ * level, this function forces the inclusion of higher level nodes. For
+ * example, given this tree:
+ *
+ * / {
+ * testing {
+ * }
+ * }
+ *
+ * If we decide to include testing then we need the root node to have a valid
+ * tree. This function adds those regions.
+ *
+ * @info: State information
+ * @depth: Current stack depth
+ */
+static int fdt_include_supernodes(struct fdt_region_state *info, int depth)
+{
+ int base = fdt_off_dt_struct(info->fdt);
+ int start, stop_at;
+ int i;
+
+ /*
+ * Work down the stack looking for supernodes that we didn't include.
+ * The algortihm here is actually pretty simple, since we know that
+ * no previous subnode had to include these nodes, or if it did, we
+ * marked them as included (on the stack) already.
+ */
+ for (i = 0; i <= depth; i++) {
+ if (!info->stack[i].included) {
+ start = info->stack[i].offset;
+
+ /* Add the FDT_BEGIN_NODE tag of this supernode */
+ fdt_next_tag(info->fdt, start, &stop_at);
+ if (fdt_add_region(info, base + start,
+ stop_at - start))
+ return -1;
+
+ /* Remember that this supernode is now included */
+ info->stack[i].included = 1;
+ info->can_merge = 1;
+ }
+
+ /* Force (later) generation of the FDT_END_NODE tag */
+ if (!info->stack[i].want)
+ info->stack[i].want = WANT_NODES_ONLY;
+ }
+
+ return 0;
+}
+
+enum {
+ FDT_DONE_NOTHING,
+ FDT_DONE_MEM_RSVMAP,
+ FDT_DONE_STRUCT,
+ FDT_DONE_END,
+ FDT_DONE_STRINGS,
+ FDT_DONE_ALL,
+};
+
+int fdt_first_region(const void *fdt,
+ int (*h_include)(void *priv, const void *fdt, int offset,
+ int type, const char *data, int size),
+ void *priv, struct fdt_region *region,
+ char *path, int path_len, int flags,
+ struct fdt_region_state *info)
+{
+ struct fdt_region_ptrs *p = &info->ptrs;
+
+ /* Set up our state */
+ info->fdt = fdt;
+ info->can_merge = 1;
+ info->max_regions = 1;
+ info->start = -1;
+ p->want = WANT_NOTHING;
+ p->end = path;
+ *p->end = '\0';
+ p->nextoffset = 0;
+ p->depth = -1;
+ p->done = FDT_DONE_NOTHING;
+
+ return fdt_next_region(fdt, h_include, priv, region,
+ path, path_len, flags, info);
+}
+
+/*
+ * Theory of operation
+ *
+ *
+ *
+
+Note: in this description 'included' means that a node (or other part of
+the tree) should be included in the region list, i.e. it will have a region
+which covers its part of the tree.
+
+This function maintains some state from the last time it is called. It
+checks the next part of the tree that it is supposed to look at
+(p.nextoffset) to see if that should be included or not. When it finds
+something to include, it sets info->start to its offset. This marks the
+start of the region we want to include.
+
+Once info->start is set to the start (i.e. not -1), we continue scanning
+until we find something that we don't want included. This will be the end
+of a region. At this point we can close off the region and add it to the
+list. So we do so, and reset info->start to -1.
+
+One complication here is that we want to merge regions. So when we come to
+add another region later, we may in fact merge it with the previous one if
+one ends where the other starts.
+
+The function fdt_add_region() will return -1 if it fails to add the region,
+because we already have a region ready to be returned, and the new one
+cannot be merged in with it. In this case, we must return the region we
+found, and wait for another call to this function. When it comes, we will
+repeat the processing of the tag and again try to add a region. This time it
+will succeed.
+
+The current state of the pointers (stack, offset, etc.) is maintained in
+a ptrs member. At the start of every loop iteration we make a copy of it.
+The copy is then updated as the tag is processed. Only if we get to the end
+of the loop iteration (and successfully call fdt_add_region() if we need
+to) can we commit the changes we have made to these pointers. For example,
+if we see an FDT_END_NODE tag we will decrement the depth value. But if we
+need to add a region for this tag (let's say because the previous tag is
+included and this FDT_END_NODE tag is not included) then we will only commit
+the result if we were able to add the region. That allows us to retry again
+next time.
+
+We keep track of a variable called 'want' which tells us what we want to
+include when there is no specific information provided by the h_include
+function for a particular property. This basically handles the inclusion of
+properties which are pulled in by virtue of the node they are in. So if you
+include a node, its properties are also included. In this case 'want' will
+be WANT_NODES_AND_PROPS. The FDT_REG_DIRECT_SUBNODES feature also makes use
+of 'want'. While we are inside the subnode, 'want' will be set to
+WANT_NODES_ONLY, so that only the subnode's FDT_BEGIN_NODE and FDT_END_NODE
+tags will be included, and properties will be skipped. If WANT_NOTHING is
+selected, then we will just rely on what the h_include() function tells us.
+
+Using 'want' we work out 'include', which tells us whether this current tag
+should be included or not. As you can imagine, if the value of 'include'
+changes, that means we are on a boundary between nodes to include and nodes
+to exclude. At this point we either close off a previous region and add it
+to the list, or mark the start of a new region.
+
+Apart from the nodes, we have mem_rsvmap, the FDT_END tag and the string
+list. Each of these dealt with as a whole (i.e. we create a region for each
+if it is to be included). For mem_rsvmap we don't allow it to merge with
+the first struct region. For the stringlist we don't allow it to merge with
+the last struct region (which contains at minimum the FDT_END tag).
+*/
+int fdt_next_region(const void *fdt,
+ int (*h_include)(void *priv, const void *fdt, int offset,
+ int type, const char *data, int size),
+ void *priv, struct fdt_region *region,
+ char *path, int path_len, int flags,
+ struct fdt_region_state *info)
+{
+ int base = fdt_off_dt_struct(fdt);
+ const char *str;
+
+ info->region = region;
+ info->count = 0;
+ if (info->ptrs.done < FDT_DONE_MEM_RSVMAP &&
+ (flags & FDT_REG_ADD_MEM_RSVMAP)) {
+ /* Add the memory reserve map into its own region */
+ if (fdt_add_region(info, fdt_off_mem_rsvmap(fdt),
+ fdt_off_dt_struct(fdt) - fdt_off_mem_rsvmap(fdt)))
+ return 0;
+ info->can_merge = 0; /* Don't allow merging with this */
+ info->ptrs.done = FDT_DONE_MEM_RSVMAP;
+ }
+
+ /*
+ * Work through the tags one by one, deciding whether each needs to
+ * be included or not. We set the variable 'include' to indicate our
+ * decision. 'want' is used to track what we want to include - it
+ * allows us to pick up all the properties (and/or subnode tags) of
+ * a node.
+ */
+ while (info->ptrs.done < FDT_DONE_STRUCT) {
+ const struct fdt_property *prop;
+ struct fdt_region_ptrs p;
+ const char *name;
+ int include = 0;
+ int stop_at = 0;
+ uint32_t tag;
+ int offset;
+ int val;
+ int len;
+
+ /*
+ * Make a copy of our pointers. If we make it to the end of
+ * this block then we will commit them back to info->ptrs.
+ * Otherwise we can try again from the same starting state
+ * next time we are called.
+ */
+ p = info->ptrs;
+
+ /*
+ * Find the tag, and the offset of the next one. If we need to
+ * stop including tags, then by default we stop *after*
+ * including the current tag
+ */
+ offset = p.nextoffset;
+ tag = fdt_next_tag(fdt, offset, &p.nextoffset);
+ stop_at = p.nextoffset;
+
+ switch (tag) {
+ case FDT_PROP:
+ stop_at = offset;
+ prop = fdt_get_property_by_offset(fdt, offset, NULL);
+ str = fdt_string(fdt, fdt32_to_cpu(prop->nameoff));
+ val = h_include(priv, fdt, offset, FDT_IS_PROP, str,
+ strlen(str) + 1);
+ if (val == -1) {
+ include = p.want >= WANT_NODES_AND_PROPS;
+ } else {
+ include = val;
+ /*
+ * Make sure we include the } for this block.
+ * It might be more correct to have this done
+ * by the call to fdt_include_supernodes() in
+ * the case where it adds the node we are
+ * currently in, but this is equivalent.
+ */
+ if ((flags & FDT_REG_SUPERNODES) && val &&
+ !p.want)
+ p.want = WANT_NODES_ONLY;
+ }
+
+ /* Value grepping is not yet supported */
+ break;
+
+ case FDT_NOP:
+ include = p.want >= WANT_NODES_AND_PROPS;
+ stop_at = offset;
+ break;
+
+ case FDT_BEGIN_NODE:
+ p.depth++;
+ if (p.depth == FDT_MAX_DEPTH)
+ return -FDT_ERR_TOODEEP;
+ name = fdt_get_name(fdt, offset, &len);
+ if (p.end - path + 2 + len >= path_len)
+ return -FDT_ERR_NOSPACE;
+
+ /* Build the full path of this node */
+ if (p.end != path + 1)
+ *p.end++ = '/';
+ strcpy(p.end, name);
+ p.end += len;
+ info->stack[p.depth].want = p.want;
+ info->stack[p.depth].offset = offset;
+
+ /*
+ * If we are not intending to include this node unless
+ * it matches, make sure we stop *before* its tag.
+ */
+ if (p.want == WANT_NODES_ONLY ||
+ !(flags & (FDT_REG_DIRECT_SUBNODES |
+ FDT_REG_ALL_SUBNODES))) {
+ stop_at = offset;
+ p.want = WANT_NOTHING;
+ }
+ val = h_include(priv, fdt, offset, FDT_IS_NODE, path,
+ p.end - path + 1);
+
+ /* Include this if requested */
+ if (val) {
+ p.want = (flags & FDT_REG_ALL_SUBNODES) ?
+ WANT_ALL_NODES_AND_PROPS :
+ WANT_NODES_AND_PROPS;
+ }
+
+ /* If not requested, decay our 'p.want' value */
+ else if (p.want) {
+ if (p.want != WANT_ALL_NODES_AND_PROPS)
+ p.want--;
+
+ /* Not including this tag, so stop now */
+ } else {
+ stop_at = offset;
+ }
+
+ /*
+ * Decide whether to include this tag, and update our
+ * stack with the state for this node
+ */
+ include = p.want;
+ info->stack[p.depth].included = include;
+ break;
+
+ case FDT_END_NODE:
+ include = p.want;
+ if (p.depth < 0)
+ return -FDT_ERR_BADSTRUCTURE;
+
+ /*
+ * If we don't p.want this node, stop right away, unless
+ * we are including subnodes
+ */
+ if (!p.want && !(flags & FDT_REG_DIRECT_SUBNODES))
+ stop_at = offset;
+ p.want = info->stack[p.depth].want;
+ p.depth--;
+ while (p.end > path && *--p.end != '/')
+ ;
+ *p.end = '\0';
+ break;
+
+ case FDT_END:
+ /* We always include the end tag */
+ include = 1;
+ p.done = FDT_DONE_STRUCT;
+ break;
+ }
+
+ /* If this tag is to be included, mark it as region start */
+ if (include && info->start == -1) {
+ /* Include any supernodes required by this one */
+ if (flags & FDT_REG_SUPERNODES) {
+ if (fdt_include_supernodes(info, p.depth))
+ return 0;
+ }
+ info->start = offset;
+ }
+
+ /*
+ * If this tag is not to be included, finish up the current
+ * region.
+ */
+ if (!include && info->start != -1) {
+ if (fdt_add_region(info, base + info->start,
+ stop_at - info->start))
+ return 0;
+ info->start = -1;
+ info->can_merge = 1;
+ }
+
+ /* If we have made it this far, we can commit our pointers */
+ info->ptrs = p;
+ }
+
+ /* Add a region for the END tag and a separate one for string table */
+ if (info->ptrs.done < FDT_DONE_END) {
+ if (info->ptrs.nextoffset != fdt_size_dt_struct(fdt))
+ return -FDT_ERR_BADSTRUCTURE;
+
+ if (fdt_add_region(info, base + info->start,
+ info->ptrs.nextoffset - info->start))
+ return 0;
+ info->ptrs.done++;
+ }
+ if (info->ptrs.done < FDT_DONE_STRINGS &&
+ (flags & FDT_REG_ADD_STRING_TAB)) {
+ info->can_merge = 0;
+ if (fdt_off_dt_strings(fdt) < base + info->ptrs.nextoffset)
+ return -FDT_ERR_BADLAYOUT;
+ if (fdt_add_region(info, fdt_off_dt_strings(fdt),
+ fdt_size_dt_strings(fdt)))
+ return 0;
+ info->ptrs.done++;
+ }
+
+ return info->count > 0 ? 0 : -FDT_ERR_NOTFOUND;
+}
diff --git a/libfdt/libfdt.h b/libfdt/libfdt.h
index 02baa84..0a401bf 100644
--- a/libfdt/libfdt.h
+++ b/libfdt/libfdt.h
@@ -116,7 +116,12 @@
* Should never be returned, if it is, it indicates a bug in
* libfdt itself. */
-#define FDT_ERR_MAX 13
+#define FDT_ERR_TOODEEP 14
+ /* FDT_ERR_TOODEEP: The depth of a node has exceeded the internal
+ * libfdt limit. This can happen if you have more than
+ * FDT_MAX_DEPTH nested nodes. */
+
+#define FDT_ERR_MAX 14
/**********************************************************************/
/* Low-level functions (you probably don't need these) */
@@ -1511,4 +1516,214 @@ int fdt_del_node(void *fdt, int nodeoffset);
const char *fdt_strerror(int errval);
+struct fdt_region {
+ int offset;
+ int size;
+};
+
+/*
+ * Flags for fdt_find_regions()
+ *
+ * Add a region for the string table (always the last region)
+ */
+#define FDT_REG_ADD_STRING_TAB (1 << 0)
+
+/*
+ * Add all supernodes of a matching node/property, useful for creating a
+ * valid subset tree
+ */
+#define FDT_REG_SUPERNODES (1 << 1)
+
+/* Add the FDT_BEGIN_NODE tags of subnodes, including their names */
+#define FDT_REG_DIRECT_SUBNODES (1 << 2)
+
+/* Add all subnodes of a matching node */
+#define FDT_REG_ALL_SUBNODES (1 << 3)
+
+/* Add a region for the mem_rsvmap table (always the first region) */
+#define FDT_REG_ADD_MEM_RSVMAP (1 << 4)
+
+/* Indicates what an fdt part is (node, property, value) */
+#define FDT_IS_NODE (1 << 0)
+#define FDT_IS_PROP (1 << 1)
+#define FDT_IS_VALUE (1 << 2) /* not supported */
+#define FDT_IS_COMPAT (1 << 3) /* used internally */
+
+#define FDT_IS_ANY 15
+
+/* We set a reasonable limit on the number of nested nodes */
+#define FDT_MAX_DEPTH 32
+
+/* Decribes what we want to include from the current tag */
+enum want_t {
+ WANT_NOTHING,
+ WANT_NODES_ONLY, /* No proporties */
+ WANT_NODES_AND_PROPS, /* Everything for one level */
+ WANT_ALL_NODES_AND_PROPS /* Everything for all levels */
+};
+
+/* Keeps track of the state at parent nodes */
+struct fdt_subnode_stack {
+ int offset; /* Offset of node */
+ enum want_t want; /* The 'want' value here */
+ int included; /* 1 if we included this node, 0 if not */
+};
+
+struct fdt_region_ptrs {
+ int depth; /* Current tree depth */
+ int done; /* What we have completed scanning */
+ enum want_t want; /* What we are currently including */
+ char *end; /* Pointer to end of full node path */
+ int nextoffset; /* Next node offset to check */
+};
+
+/* The state of our finding algortihm */
+struct fdt_region_state {
+ struct fdt_subnode_stack stack[FDT_MAX_DEPTH]; /* node stack */
+ struct fdt_region *region; /* Contains list of regions found */
+ int count; /* Numnber of regions found */
+ const void *fdt; /* FDT blob */
+ int max_regions; /* Maximum regions to find */
+ int can_merge; /* 1 if we can merge with previous region */
+ int start; /* Start position of current region */
+ struct fdt_region_ptrs ptrs; /* Pointers for what we are up to */
+};
+
+/**
+ * fdt_first_region() - find regions in device tree
+ *
+ * Given a nodes and properties to include and properties to exclude, find
+ * the regions of the device tree which describe those included parts.
+ *
+ * The use for this function is twofold. Firstly it provides a convenient
+ * way of performing a structure-aware grep of the tree. For example it is
+ * possible to grep for a node and get all the properties associated with
+ * that node. Trees can be subsetted easily, by specifying the nodes that
+ * are required, and then writing out the regions returned by this function.
+ * This is useful for small resource-constrained systems, such as boot
+ * loaders, which want to use an FDT but do not need to know about all of
+ * it.
+ *
+ * Secondly it makes it easy to hash parts of the tree and detect changes.
+ * The intent is to get a list of regions which will be invariant provided
+ * those parts are invariant. For example, if you request a list of regions
+ * for all nodes but exclude the property "data", then you will get the
+ * same region contents regardless of any change to "data" properties.
+ *
+ * This function can be used to produce a byte-stream to send to a hashing
+ * function to verify that critical parts of the FDT have not changed.
+ * Note that semantically null changes in order could still cause false
+ * hash misses. Such reordering might happen if the tree is regenerated
+ * from source, and nodes are reordered (the bytes-stream will be emitted
+ * in a different order and mnay hash functions will detect this). However
+ * if an existing tree is modified using libfdt functions, such as
+ * fdt_add_subnode() and fdt_setprop(), then this problem is avoided.
+ *
+ * The nodes/properties to include/exclude are defined by a function
+ * provided by the caller. This function is called for each node and
+ * property, and must return:
+ *
+ * 0 - to exclude this part
+ * 1 - to include this part
+ * -1 - for FDT_IS_PROP only: no information is available, so include
+ * if its containing node is included
+ *
+ * The last case is only used to deal with properties. Often a property is
+ * included if its containing node is included - this is the case where
+ * -1 is returned.. However if the property is specifically required to be
+ * included/excluded, then 0 or 1 can be returned. Note that including a
+ * property when the FDT_REG_SUPERNODES flag is given will force its
+ * containing node to be included since it is not valid to have a property
+ * that is not in a node.
+ *
+ * Using the information provided, the inclusion of a node can be controlled
+ * either by a node name or its compatible string, or any other property
+ * that the function can determine.
+ *
+ * As an example, including node "/" means to include the root node and all
+ * root properties. A flag provides a way of also including supernodes (of
+ * which there is none for the root node), and another flag includes
+ * immediate subnodes, so in this case we would get the FDT_BEGIN_NODE and
+ * FDT_END_NODE of all subnodes of /.
+ *
+ * The subnode feature helps in a hashing situation since it prevents the
+ * root node from changing at all. Any change to non-excluded properties,
+ * names of subnodes or number of subnodes would be detected.
+ *
+ * When used with FITs this provides the ability to hash and sign parts of
+ * the FIT based on different configurations in the FIT. Then it is
+ * impossible to change anything about that configuration (include images
+ * attached to the configuration), but it may be possible to add new
+ * configurations, new images or new signatures within the existing
+ * framework.
+ *
+ * Adding new properties to a device tree may result in the string table
+ * being extended (if the new property names are different from those
+ * already added). This function can optionally include a region for
+ * the string table so that this can be part of the hash too. This is always
+ * the last region.
+ *
+ * The FDT also has a mem_rsvmap table which can also be included, and is
+ * always the first region if so.
+ *
+ * The device tree header is not included in the region list. Since the
+ * contents of the FDT are changing (shrinking, often), the caller will need
+ * to regenerate the header anyway.
+ *
+ * @fdt: Device tree to check
+ * @h_include: Function to call to determine whether to include a part or
+ * not:
+ *
+ * @priv: Private pointer as passed to fdt_find_regions()
+ * @fdt: Pointer to FDT blob
+ * @offset: Offset of this node / property
+ * @type: Type of this part, FDT_IS_...
+ * @data: Pointer to data (node name, property name, compatible
+ * string, value (not yet supported)
+ * @size: Size of data, or 0 if none
+ * @return 0 to exclude, 1 to include, -1 if no information is
+ * available
+ * @priv: Private pointer passed to h_include
+ * @region: Returns list of regions, sorted by offset
+ * @max_regions: Maximum length of region list
+ * @path: Pointer to a temporary string for the function to use for
+ * building path names
+ * @path_len: Length of path, must be large enough to hold the longest
+ * path in the tree
+ * @flags: Various flags that control the region algortihm, see
+ * FDT_REG_...
+ * @return number of regions in list. If this is >max_regions then the
+ * region array was exhausted. You should increase max_regions and try
+ * the call again. Only the first max_regions elements are available in the
+ * array.
+ *
+ * On error a -ve value is return, which can be:
+ *
+ * -FDT_ERR_BADSTRUCTURE (too deep or more END tags than BEGIN tags
+ * -FDT_ERR_BADLAYOUT
+ * -FDT_ERR_NOSPACE (path area is too small)
+ */
+int fdt_first_region(const void *fdt,
+ int (*h_include)(void *priv, const void *fdt, int offset,
+ int type, const char *data, int size),
+ void *priv, struct fdt_region *region,
+ char *path, int path_len, int flags,
+ struct fdt_region_state *info);
+
+/** fdt_next_region() - find next region
+ *
+ * See fdt_first_region() for full description. This function finds the
+ * next region according to the provided parameters, which must be the same
+ * as passed to fdt_first_region().
+ *
+ * This function can additionally return -FDT_ERR_NOTFOUND when there are no
+ * more regions
+ */
+int fdt_next_region(const void *fdt,
+ int (*h_include)(void *priv, const void *fdt, int offset,
+ int type, const char *data, int size),
+ void *priv, struct fdt_region *region,
+ char *path, int path_len, int flags,
+ struct fdt_region_state *info);
+
#endif /* _LIBFDT_H */
diff --git a/tests/.gitignore b/tests/.gitignore
index bb5e33a..4dc6466 100644
--- a/tests/.gitignore
+++ b/tests/.gitignore
@@ -41,6 +41,7 @@ tmp.*
/phandle_format
/propname_escapes
/references
+/region_tree
/root_node
/rw_tree1
/set_name
diff --git a/tests/Makefile.tests b/tests/Makefile.tests
index dafb618..636a242 100644
--- a/tests/Makefile.tests
+++ b/tests/Makefile.tests
@@ -21,7 +21,8 @@ LIB_TESTS_L = get_mem_rsv \
add_subnode_with_nops path_offset_aliases \
utilfdt_test \
integer-expressions \
- subnode_iterate
+ subnode_iterate \
+ region_tree
LIB_TESTS = $(LIB_TESTS_L:%=$(TESTS_PREFIX)%)
LIBTREE_TESTS_L = truncated_property
diff --git a/tests/region_tree.c b/tests/region_tree.c
new file mode 100644
index 0000000..2ca8c98
--- /dev/null
+++ b/tests/region_tree.c
@@ -0,0 +1,352 @@
+/*
+ * hash_tree - Testcase for fdt_find_regions()
+ *
+ * Copyright (c) 2013, Google Inc.
+ *
+ * Copyright (C) 2006 David Gibson, IBM Corporation.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public License
+ * as published by the Free Software Foundation; either version 2.1 of
+ * the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <ctype.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include <libfdt_env.h>
+#include <fdt.h>
+#include <libfdt.h>
+
+#include "tests.h"
+#include "testdata.h"
+
+#define SPACE 65536
+
+#define CHECK(code) \
+ { \
+ err = (code); \
+ if (err) \
+ FAIL(#code ": %s", fdt_strerror(err)); \
+ }
+
+/*
+ * Regions we expect to see returned from fdt_find_regions(). We build up a
+ * list of these as we make the tree, then check the results of
+ * fdt_find_regions() once we are done.
+ */
+static struct fdt_region expect[20];
+
+/* Number of expected regions */
+int expect_count;
+
+/* Mark the start of a new region */
+static void start(void *fdt)
+{
+ expect[expect_count].offset = fdt_size_dt_struct(fdt);
+ verbose_printf("[%d: %x ", expect_count,
+ fdt_off_dt_struct(fdt) + expect[expect_count].offset);
+}
+
+/* Mark the end of a region */
+static void stop(void *fdt)
+{
+ expect[expect_count].size = fdt_size_dt_struct(fdt) -
+ expect[expect_count].offset;
+ expect[expect_count].offset += fdt_off_dt_struct(fdt);
+ verbose_printf("%x]\n", expect[expect_count].offset +
+ expect[expect_count].size);
+ expect_count++;
+}
+
+/**
+ * build_tree() - Build a tree
+ *
+ * @fdt: Pointer to place to put tree, assumed to be large enough
+ * @flags: Flags to control the tree creation (FDT_REG_...)
+ * @space: Amount of space to create for later tree additions
+ *
+ * This creates a tree modelled on a U-Boot FIT image, with various nodes
+ * and properties which are useful for testing the hashing features of
+ * fdt_find_regions().
+ *
+ * See h_include() below for a list of the nodes we later search for.
+ */
+static void build_tree(void *fdt, int flags, int space)
+{
+ int direct_subnodes = flags & FDT_REG_DIRECT_SUBNODES;
+ int all_subnodes = flags & FDT_REG_ALL_SUBNODES;
+ int supernodes = flags & FDT_REG_SUPERNODES;
+ int either = !all_subnodes && (direct_subnodes || supernodes);
+ int err;
+
+ CHECK(fdt_create(fdt, SPACE));
+
+ CHECK(fdt_add_reservemap_entry(fdt, TEST_ADDR_1, TEST_SIZE_1));
+ CHECK(fdt_add_reservemap_entry(fdt, TEST_ADDR_2, TEST_SIZE_2));
+ CHECK(fdt_finish_reservemap(fdt));
+
+ /*
+ * This is the start of a new region because in the fdt_xxx_region()
+ * call, we pass "/" as one of the nodes to find.
+ */
+ start(fdt); /* region 0 */
+ CHECK(fdt_begin_node(fdt, ""));
+ CHECK(fdt_property_string(fdt, "description", "kernel image"));
+ CHECK(fdt_property_u32(fdt, "#address-cells", 1));
+
+ /* /images */
+ if (!either && !all_subnodes)
+ stop(fdt);
+ CHECK(fdt_begin_node(fdt, "images"));
+ if (either)
+ stop(fdt);
+ CHECK(fdt_property_u32(fdt, "image-prop", 1));
+
+ /* /images/kernel@1 */
+ if (!all_subnodes)
+ start(fdt); /* region 1 */
+ CHECK(fdt_begin_node(fdt, "kernel@1"));
+ CHECK(fdt_property_string(fdt, "description", "exynos kernel"));
+ stop(fdt);
+ CHECK(fdt_property_string(fdt, "data", "this is the kernel image"));
+ start(fdt); /* region 2 */
+
+ /* /images/kernel/hash@1 */
+ CHECK(fdt_begin_node(fdt, "hash@1"));
+ CHECK(fdt_property_string(fdt, "algo", "sha1"));
+ CHECK(fdt_end_node(fdt));
+
+ /* /images/kernel/hash@2 */
+ if (!direct_subnodes)
+ stop(fdt);
+ CHECK(fdt_begin_node(fdt, "hash@2"));
+ if (direct_subnodes)
+ stop(fdt);
+ CHECK(fdt_property_string(fdt, "algo", "sha1"));
+ if (direct_subnodes)
+ start(fdt); /* region 3 */
+ CHECK(fdt_end_node(fdt));
+ if (!direct_subnodes)
+ start(fdt); /* region 3 */
+
+ CHECK(fdt_end_node(fdt));
+
+ /* /images/fdt@1 */
+ CHECK(fdt_begin_node(fdt, "fdt@1"));
+ CHECK(fdt_property_string(fdt, "description", "snow FDT"));
+ if (!all_subnodes)
+ stop(fdt);
+ CHECK(fdt_property_string(fdt, "data", "FDT data"));
+ if (!all_subnodes)
+ start(fdt); /* region 4 */
+
+ /* /images/kernel/hash@1 */
+ CHECK(fdt_begin_node(fdt, "hash@1"));
+ CHECK(fdt_property_string(fdt, "algo", "sha1"));
+ CHECK(fdt_end_node(fdt));
+
+ CHECK(fdt_end_node(fdt));
+
+ if (!either && !all_subnodes)
+ stop(fdt);
+ CHECK(fdt_end_node(fdt));
+
+ /* /configurations */
+ CHECK(fdt_begin_node(fdt, "configurations"));
+ if (either)
+ stop(fdt);
+ CHECK(fdt_property_string(fdt, "default", "conf@1"));
+
+ /* /configurations/conf@1 */
+ if (!all_subnodes)
+ start(fdt); /* region 6 */
+ CHECK(fdt_begin_node(fdt, "conf@1"));
+ CHECK(fdt_property_string(fdt, "kernel", "kernel@1"));
+ CHECK(fdt_property_string(fdt, "fdt", "fdt@1"));
+ CHECK(fdt_end_node(fdt));
+ if (!all_subnodes)
+ stop(fdt);
+
+ /* /configurations/conf@2 */
+ CHECK(fdt_begin_node(fdt, "conf@2"));
+ CHECK(fdt_property_string(fdt, "kernel", "kernel@1"));
+ CHECK(fdt_property_string(fdt, "fdt", "fdt@2"));
+ CHECK(fdt_end_node(fdt));
+
+ if (either)
+ start(fdt); /* region 7 */
+ CHECK(fdt_end_node(fdt));
+ if (!either && !all_subnodes)
+ start(fdt); /* region 7 */
+
+ CHECK(fdt_end_node(fdt));
+
+ CHECK(fdt_finish(fdt));
+ stop(fdt);
+
+ /* Add in the strings */
+ if (flags & FDT_REG_ADD_STRING_TAB) {
+ expect[expect_count].offset = fdt_off_dt_strings(fdt);
+ expect[expect_count].size = fdt_size_dt_strings(fdt);
+ expect_count++;
+ }
+
+ /* Make a bit of space */
+ if (space)
+ CHECK(fdt_open_into(fdt, fdt, fdt_totalsize(fdt) + space));
+
+ verbose_printf("Completed tree, totalsize = %d\n", fdt_totalsize(fdt));
+}
+
+/**
+ * strlist_contains() - Returns 1 if a string is contained in a list
+ *
+ * @list: List of strings
+ * @count: Number of strings in list
+ * @str: String to search for
+ */
+static int strlist_contains(const char * const list[], int count,
+ const char *str)
+{
+ int i;
+
+ for (i = 0; i < count; i++)
+ if (!strcmp(list[i], str))
+ return 1;
+
+ return 0;
+}
+
+/**
+ * h_include() - Our include handler for fdt_find_regions()
+ *
+ * This is very simple - we have a list of nodes we are looking for, and
+ * one property that we want to exclude.
+ */
+static int h_include(void *priv, const void *fdt, int offset, int type,
+ const char *data, int size)
+{
+ const char * const inc[] = {
+ "/",
+ "/images/kernel@1",
+ "/images/fdt@1",
+ "/configurations/conf@1",
+ "/images/kernel@1/hash@1",
+ "/images/fdt@1/hash@1",
+ };
+
+ switch (type) {
+ case FDT_IS_NODE:
+ return strlist_contains(inc, 6, data);
+ case FDT_IS_PROP:
+ return !strcmp(data, "data") ? 0 : -1;
+ }
+
+ return 0;
+}
+
+/**
+ * check_regions() - Check that the regions are as we expect
+ *
+ * Call fdt_find_regions() and check that the results are as we expect them,
+ * matching the list of expected regions we created at the same time as
+ * the tree.
+ *
+ * @fdt: Pointer to device tree to check
+ * @flags: Flags value (FDT_REG_...)
+ * @return 0 if ok, -1 on failure
+ */
+static int check_regions(const void *fdt, int flags)
+{
+ struct fdt_region_state state;
+ struct fdt_region reg;
+ int err, ret = 0;
+ char path[1024];
+ int count;
+ int i;
+
+ ret = fdt_first_region(fdt, h_include, NULL, ®,
+ path, sizeof(path), flags, &state);
+ if (ret < 0)
+ CHECK(ret);
+
+ verbose_printf("Regions: %d\n", count);
+ for (i = 0; ; i++) {
+ struct fdt_region *exp = &expect[i];
+
+ verbose_printf("%d: %-10x %-10x\n", i, reg.offset,
+ reg.offset + reg.size);
+ if (memcmp(exp, ®, sizeof(reg))) {
+ ret = -1;
+ verbose_printf("exp: %-10x %-10x\n", exp->offset,
+ exp->offset + exp->size);
+ }
+
+ ret = fdt_next_region(fdt, h_include, NULL, ®,
+ path, sizeof(path), flags, &state);
+ if (ret < 0) {
+ if (ret == -FDT_ERR_NOTFOUND)
+ ret = 0;
+ CHECK(ret);
+ i++;
+ break;
+ }
+ }
+ verbose_printf("expect_count = %d, i=%d\n", expect_count, i);
+ if (expect_count != i)
+ FAIL();
+
+ return ret;
+}
+
+int main(int argc, char *argv[])
+{
+ const char *fname = NULL;
+ int flags = 0;
+ int space = 0;
+ void *fdt;
+
+ test_init(argc, argv);
+ if (argc < 2) {
+ verbose_printf("Usage: %s <flag value> [<space>]"
+ " [<output_fname.dtb>]", argv[0]);
+ FAIL();
+ }
+ flags = atoi(argv[1]);
+ if (argc >= 3)
+ space = atoi(argv[2]);
+ if (argc >= 4)
+ fname = argv[3];
+
+ /*
+ * Allocate space for the tree and build it, creating a list of
+ * expected regions.
+ */
+ fdt = xmalloc(SPACE);
+ build_tree(fdt, flags, space);
+
+ /* Write the tree out if required */
+ if (fname)
+ save_blob(fname, fdt);
+
+ /* Check the regions are what we expect */
+ if (check_regions(fdt, flags))
+ FAIL();
+ else
+ PASS();
+
+ return 0;
+}
diff --git a/tests/run_tests.sh b/tests/run_tests.sh
index c0a136b..8757c11 100755
--- a/tests/run_tests.sh
+++ b/tests/run_tests.sh
@@ -249,6 +249,11 @@ libfdt_tests () {
# Specific bug tests
run_test add_subnode_with_nops
+
+ # Tests for fdt_find_regions()
+ for flags in $(seq 0 15); do
+ run_test region_tree ${flags}
+ done
}
dtc_tests () {
--
1.9.1.423.g4596e3a
--
To unsubscribe from this list: send the line "unsubscribe devicetree-compiler" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [RESEND PATCH v3 2/3] Add documentation for fdtget/put
[not found] ` <1395500836-12735-1-git-send-email-sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
2014-03-22 15:07 ` [RESEND PATCH v3 1/3] libfdt: Add function to find regions in an FDT Simon Glass
@ 2014-03-22 15:07 ` Simon Glass
2014-03-22 15:07 ` [RESEND PATCH v3 3/3] Add fdtgrep to grep and subset FDTs Simon Glass
2 siblings, 0 replies; 9+ messages in thread
From: Simon Glass @ 2014-03-22 15:07 UTC (permalink / raw)
To: Devicetree Compiler; +Cc: Jon Loeliger, David Gibson, Simon Glass
The other tools have documentation so it seems reasonable to add something
here.
Signed-off-by: Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
---
Changes in v3: None
Changes in v2: None
Documentation/manual.txt | 38 ++++++++++++++++++++++++++++++++++++++
1 file changed, 38 insertions(+)
diff --git a/Documentation/manual.txt b/Documentation/manual.txt
index 65c8540..5f02426 100644
--- a/Documentation/manual.txt
+++ b/Documentation/manual.txt
@@ -658,3 +658,41 @@ The fdtdump program prints a readable version of a flat device tree file.
The syntax of the fdtdump command line is:
fdtdump <DTB-file-name>
+
+
+3) fdtget -- Get individual properties and lists from a Device Tree
+
+Ths fdtget program allows you to extract properties from nodes, and also
+list nodes and properties.
+
+The syntax of the fdtget commandline is described in the help (fdtget -h).
+Common uses are:
+
+ fdtget -ts <DTB-file-name> /node compatible
+ - Read the compatible string of a node. The -ts is optional since
+ fdtget will normally guess the type correctly.
+
+ fdtget -bx <DTB-file-name> /node bytes
+ - Read the bytes in a property and output in hex. The 'b' prefix
+ forces byte output, since otherwise fdtget will output in 32-bit
+ or 16-bit words if the property length is so-aligned.
+
+ fdtget -p <DTB-file-name> /node
+ - List the properties in a node
+
+ fdtget -l <DTB-file-name> /
+ - List the subnodes for a node (here, the root node)
+
+You can use -d to provide a default value for when the property does not
+exist.
+
+
+4) fdtput -- Write individual properties to a Device Tree
+
+Ths fdtput program allows you to write properties to nodes.
+
+The syntax of the fdtput commandline is described in the help (fdtput -h) and
+the type options are the same as fdtget. You can use -c to create a node if
+it does not exist already, and -p to create all required nodes (like
+'mkdir -p' does with directories). Unfortunately fdtput does not support
+deleting nodes or properties.
--
1.9.1.423.g4596e3a
--
To unsubscribe from this list: send the line "unsubscribe devicetree-compiler" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [RESEND PATCH v3 3/3] Add fdtgrep to grep and subset FDTs
[not found] ` <1395500836-12735-1-git-send-email-sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
2014-03-22 15:07 ` [RESEND PATCH v3 1/3] libfdt: Add function to find regions in an FDT Simon Glass
2014-03-22 15:07 ` [RESEND PATCH v3 2/3] Add documentation for fdtget/put Simon Glass
@ 2014-03-22 15:07 ` Simon Glass
2 siblings, 0 replies; 9+ messages in thread
From: Simon Glass @ 2014-03-22 15:07 UTC (permalink / raw)
To: Devicetree Compiler; +Cc: Jon Loeliger, David Gibson, Simon Glass
fdtgrep operates like a grep for device trees, working on binary .dtb
files.
fdtgrep can find nodes by name or compatible string. It can find properties
by name. Using a list of of nodes/properties/compatible strings to match,
fdtgrep creates either .dts text output containing just those matches, or
.dtb output. In the .dtb case, the output is a valid FDT which only includes
the selected nodes/properties. This is useful in resource-constrained systems
where a full FDT is too large to fit in available memory (e.g. U-Boot SPL).
fdtgrep also supports producing raw binary output, without the normal FDT
headers. This can be useful for hashing part of an FDT and making sure that
that part does not change, even if other parts do. This makes it possible to
detect an important change, while ignoring a change that is of no interest.
At its simplest fdtgrep can be used as follows;
fdtgrep /holiday tests/grep.dtb
and it produces just that node (with a skeleton to hold it):
/ {
holiday {
compatible = "ixtapa", "mexico";
weather = "sunny";
status = "okay";
};
};
The binary form of this is:
fdtgrep -n /holiday -O dtb tests/grep.dtb
which produces a valid DTB file with just the above nodes.
Various options are provided to search for properties, to ensure that
nodes/properties are not present, and to search for nodes by compatible
string.
Signed-off-by: Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
---
Changes in v3:
- Adjust help and command line processing to follow new approach
- Rename -V to -I to avoid using -V for a different purpose to other tools
- Rename -s to -e since it only 'enters' the node and does not include it all
- Add -s option to include all subnodes
- Add -f option to display offset; make -a display an absolute file address
- Add documentation on fdtgrep
Changes in v2:
- Add local fdt_find_regions() function since libfdt no longer has it
.gitignore | 1 +
Documentation/manual.txt | 87 +++++
Makefile | 4 +
Makefile.utils | 7 +
fdtgrep.c | 873 +++++++++++++++++++++++++++++++++++++++++++++++
tests/grep.dts | 23 ++
tests/run_tests.sh | 356 ++++++++++++++++++-
tests/tests.sh | 1 +
8 files changed, 1351 insertions(+), 1 deletion(-)
create mode 100644 fdtgrep.c
create mode 100644 tests/grep.dts
diff --git a/.gitignore b/.gitignore
index 545b899..9d3a550 100644
--- a/.gitignore
+++ b/.gitignore
@@ -12,6 +12,7 @@ lex.yy.c
/convert-dtsv0
/version_gen.h
/fdtget
+/fdtgrep
/fdtput
/patches
/.pc
diff --git a/Documentation/manual.txt b/Documentation/manual.txt
index 5f02426..77cbc38 100644
--- a/Documentation/manual.txt
+++ b/Documentation/manual.txt
@@ -696,3 +696,90 @@ the type options are the same as fdtget. You can use -c to create a node if
it does not exist already, and -p to create all required nodes (like
'mkdir -p' does with directories). Unfortunately fdtput does not support
deleting nodes or properties.
+
+
+5) fdtgrep -- Extract portions of a Device Tree and output them
+
+Ths fdtgrep program allows you to 'grep' a Device Tree file in a structured
+way. The output of fdtgrep is either another Device Tree file or a text file,
+perhaps with some pieces omitted.
+
+This is useful in a few situations:
+
+ - Finding a node or property in a device tree and displaying it along
+ with its surrounding context. This is helpful since some files are
+ quite large.
+ - Creating a smaller device tree which omits some portions. For example
+ a full Linux kernel device tree may be cut down for use by a simple
+ boot loader (perhaps removing the pinmux information).
+
+The syntax of the fdtgrep commandline is described in the help (fdtgrep -h)
+and there are many options. Some common uses are as follows:
+
+ fdtgrep -s -n /node <DTB-file-name>
+ - Output just a node and its subnodes
+
+ fdtgrep -s -n /node -o out.dtb -O dtb <DTB-file-name>
+ - Same but output as a binary Device Tree
+
+ fdtgrep -s -N /node <DTB-file-name>
+ - Output everything except the given node
+
+ fdtgrep -a -n /compatible -n /aliases <DTB-file-name>
+ - Output compatible and alias nodes
+
+ fdtgrep -s /node -O bin <DTB-file-name> | sha1sum
+ - Take the sha1sum of just the portion of the Device Tree occupied
+ by the /node node. This could be compared with the same node from
+ another file perhaps, to see if they match.
+ - You can add -tme to produce a valid Device Tree including header,
+ memreserve table and string table.
+
+ fdtgrep -A /node <DTB-file-name>
+ - Output just a node and its subnodes
+
+ fdtgrep -f /chosen spi4 <DTB-file-name>
+ - Output nodes/properties/compatibles strings which match /chosen and
+ spi4. Add the hex offset on the left of each. Note that -g is the
+ default parameter type, so this equivalent to:
+ fdtgrep -f -g /chosen -g spi4 <DTB-file-name>
+
+ fdtgrep -a /chosen spi4 <DTB-file-name>
+ - Similar but use absolute file offset. This allows to to find nodes
+ and properties in a file with a hex dumper.
+
+ fdtgrep -a /chosen spi4 <DTB-file-name>
+ - Similar but use absolute file offset. This allows to to find nodes
+ and properties in a file with a hex dumper.
+
+ fdtgrep -A /chosen spi4 <DTB-file-name>
+ - Output everything, but colour the nodes and properties which match
+ /chosen and spi4 green, and everything else red.
+
+ fdtgrep -Ad /chosen spi4 <DTB-file-name>
+ - Similar but use + and - to indicate included and excluded lines.
+
+ fdtgrep -Adv /chosen spi4 <DTB-file-name>
+ - Invert the above (-v operates similarly to -v with standard grep)
+
+ fdtgrep -c google,cros-ec-keyb <DTB-file-name>
+ - Show the node with the given compatible string
+
+ fdtgrep -n x -p compatible <DTB-file-name>
+ - List all nodes and their compatible strings. The '-n x' drops all
+ nodes not called 'x', which his all of them. If you want to list
+ nodes without a compatible as well, then omit this.
+
+Note you can use:
+ -n/N to include/exclude nodes
+ -p/N to include/exclude properties
+ -c/C to include/exclude nodes which have a particular compatible string
+ -g/G to include/exclude any of the above (global search)
+
+ Note it is not permitted to use a positive/negative search of the same
+ type at the same time. You can do this in two steps, such as:
+
+ ./fdtgrep -n x -p compatible x -O dtb |./fdtgrep - -n /serial@12C10000
+
+ but it is hard to see why this would be useful. Unfortunately fdtgrep
+ does not support wildcard matching or regular expressions.
diff --git a/Makefile b/Makefile
index 962f94e..0342081 100644
--- a/Makefile
+++ b/Makefile
@@ -112,6 +112,7 @@ BIN += dtc
BIN += fdtdump
BIN += fdtget
BIN += fdtput
+BIN += fdtgrep
SCRIPTS = dtdiff
@@ -124,6 +125,7 @@ ifneq ($(DEPTARGETS),)
-include $(FDTDUMP_OBJS:%.o=%.d)
-include $(FDTGET_OBJS:%.o=%.d)
-include $(FDTPUT_OBJS:%.o=%.d)
+-include $(FDTGREP_OBJS:%.o=%.d)
endif
@@ -188,6 +190,7 @@ fdtget: $(FDTGET_OBJS) $(LIBFDT_archive)
fdtput: $(FDTPUT_OBJS) $(LIBFDT_archive)
+fdtgrep: $(FDTGREP_OBJS) $(LIBFDT_archive)
#
# Testsuite rules
@@ -198,6 +201,7 @@ TESTS_BIN += dtc
TESTS_BIN += convert-dtsv0
TESTS_BIN += fdtput
TESTS_BIN += fdtget
+TESTS_BIN += fdtgrep
include tests/Makefile.tests
diff --git a/Makefile.utils b/Makefile.utils
index 48ece49..f2a7001 100644
--- a/Makefile.utils
+++ b/Makefile.utils
@@ -22,3 +22,10 @@ FDTPUT_SRCS = \
util.c
FDTPUT_OBJS = $(FDTPUT_SRCS:%.c=%.o)
+
+
+FDTGREP_SRCS = \
+ fdtgrep.c \
+ util.c
+
+FDTGREP_OBJS = $(FDTGREP_SRCS:%.c=%.o)
diff --git a/fdtgrep.c b/fdtgrep.c
new file mode 100644
index 0000000..010b715
--- /dev/null
+++ b/fdtgrep.c
@@ -0,0 +1,873 @@
+/*
+ * Copyright (c) 2013, Google Inc.
+ *
+ * Perform a grep of an FDT either displaying the source subset or producing
+ * a new .dtb subset which can be used as required.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ */
+
+#include <assert.h>
+#include <ctype.h>
+#include <getopt.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#include <libfdt.h>
+#include <libfdt_internal.h>
+#include "util.h"
+
+/* Define DEBUG to get some debugging output on stderr */
+#ifdef DEBUG
+#define debug(a, b...) fprintf(stderr, a, ## b)
+#else
+#define debug(a, b...)
+#endif
+
+/* A linked list of values we are grepping for */
+struct value_node {
+ int type; /* Types this value matches (FDT_IS... mask) */
+ int include; /* 1 to include matches, 0 to exclude */
+ const char *string; /* String to match */
+ struct value_node *next; /* Pointer to next node, or NULL */
+};
+
+/* Output formats we support */
+enum output_t {
+ OUT_DTS, /* Device tree source */
+ OUT_DTB, /* Valid device tree binary */
+ OUT_BIN, /* Fragment of .dtb, for hashing */
+};
+
+/* Holds information which controls our output and options */
+struct display_info {
+ enum output_t output; /* Output format */
+ int all; /* Display all properties/nodes */
+ int colour; /* Display output in ANSI colour */
+ int region_list; /* Output a region list */
+ int flags; /* Flags (FDT_REG_...) */
+ int list_strings; /* List strings in string table */
+ int show_offset; /* Show offset */
+ int show_addr; /* Show address */
+ int header; /* Output an FDT header */
+ int diff; /* Show +/- diff markers */
+ int show_dts_version; /* Put '/dts-v1/;' on the first line */
+ int types_inc; /* Mask of types that we include (FDT_IS...) */
+ int types_exc; /* Mask of types that we exclude (FDT_IS...) */
+ int invert; /* Invert polarity of match */
+ struct value_node *value_head; /* List of values to match */
+ const char *output_fname; /* Output filename */
+ FILE *fout; /* File to write dts/dtb output */
+};
+
+static void report_error(const char *where, int err)
+{
+ fprintf(stderr, "Error at '%s': %s\n", where, fdt_strerror(err));
+}
+
+/* Supported ANSI colours */
+enum {
+ COL_BLACK,
+ COL_RED,
+ COL_GREEN,
+ COL_YELLOW,
+ COL_BLUE,
+ COL_MAGENTA,
+ COL_CYAN,
+ COL_WHITE,
+
+ COL_NONE = -1,
+};
+
+/**
+ * print_ansi_colour() - Print out the ANSI sequence for a colour
+ *
+ * @fout: Output file
+ * @col: Colour to output (COL_...), or COL_NONE to reset colour
+ */
+static void print_ansi_colour(FILE *fout, int col)
+{
+ if (col == COL_NONE)
+ fprintf(fout, "\033[0m");
+ else
+ fprintf(fout, "\033[1;%dm", col + 30);
+}
+
+
+/**
+ * value_add() - Add a new value to our list of things to grep for
+ *
+ * @disp: Display structure, holding info about our options
+ * @headp: Pointer to header pointer of list
+ * @type: Type of this value (FDT_IS_...)
+ * @include: 1 if we want to include matches, 0 to exclude
+ * @str: String value to match
+ */
+static int value_add(struct display_info *disp, struct value_node **headp,
+ int type, int include, const char *str)
+{
+ struct value_node *node;
+
+ /*
+ * Keep track of which types we are excluding/including. We don't
+ * allow both including and excluding things, because it doesn't make
+ * sense. 'Including' means that everything not mentioned is
+ * excluded. 'Excluding' means that everything not mentioned is
+ * included. So using the two together would be meaningless.
+ */
+ if (include)
+ disp->types_inc |= type;
+ else
+ disp->types_exc |= type;
+ if (disp->types_inc & disp->types_exc & type) {
+ fprintf(stderr, "Cannot use both include and exclude"
+ " for '%s'\n", str);
+ return -1;
+ }
+
+ str = strdup(str);
+ node = malloc(sizeof(*node));
+ if (!str || !node) {
+ fprintf(stderr, "Out of memory\n");
+ return -1;
+ }
+ node->next = *headp;
+ node->type = type;
+ node->include = include;
+ node->string = str;
+ *headp = node;
+
+ return 0;
+}
+
+/**
+ * display_fdt_by_regions() - Display regions of an FDT source
+ *
+ * This dumps an FDT as source, but only certain regions of it. This is the
+ * final stage of the grep - we have a list of regions we want to display,
+ * and this function displays them.
+ *
+ * @disp: Display structure, holding info about our options
+ * @blob: FDT blob to display
+ * @region: List of regions to display
+ * @count: Number of regions
+ */
+static int display_fdt_by_regions(struct display_info *disp, const void *blob,
+ struct fdt_region region[], int count)
+{
+ struct fdt_region *reg = region, *reg_end = region + count;
+ uint32_t off_mem_rsvmap = fdt_off_mem_rsvmap(blob);
+ int base = fdt_off_dt_struct(blob);
+ int version = fdt_version(blob);
+ int offset, nextoffset;
+ int tag, depth, shift;
+ FILE *f = disp->fout;
+ uint64_t addr, size;
+ int in_region;
+ int file_ofs;
+ int i;
+
+ if (disp->show_dts_version)
+ fprintf(f, "/dts-v1/;\n");
+
+ if (disp->header) {
+ fprintf(f, "// magic:\t\t0x%x\n", fdt_magic(blob));
+ fprintf(f, "// totalsize:\t\t0x%x (%d)\n", fdt_totalsize(blob),
+ fdt_totalsize(blob));
+ fprintf(f, "// off_dt_struct:\t0x%x\n",
+ fdt_off_dt_struct(blob));
+ fprintf(f, "// off_dt_strings:\t0x%x\n",
+ fdt_off_dt_strings(blob));
+ fprintf(f, "// off_mem_rsvmap:\t0x%x\n", off_mem_rsvmap);
+ fprintf(f, "// version:\t\t%d\n", version);
+ fprintf(f, "// last_comp_version:\t%d\n",
+ fdt_last_comp_version(blob));
+ if (version >= 2) {
+ fprintf(f, "// boot_cpuid_phys:\t0x%x\n",
+ fdt_boot_cpuid_phys(blob));
+ }
+ if (version >= 3) {
+ fprintf(f, "// size_dt_strings:\t0x%x\n",
+ fdt_size_dt_strings(blob));
+ }
+ if (version >= 17) {
+ fprintf(f, "// size_dt_struct:\t0x%x\n",
+ fdt_size_dt_struct(blob));
+ }
+ fprintf(f, "\n");
+ }
+
+ if (disp->flags & FDT_REG_ADD_MEM_RSVMAP) {
+ const struct fdt_reserve_entry *p_rsvmap;
+
+ p_rsvmap = (const struct fdt_reserve_entry *)
+ ((const char *)blob + off_mem_rsvmap);
+ for (i = 0; ; i++) {
+ addr = fdt64_to_cpu(p_rsvmap[i].address);
+ size = fdt64_to_cpu(p_rsvmap[i].size);
+ if (addr == 0 && size == 0)
+ break;
+
+ fprintf(f, "/memreserve/ %llx %llx;\n",
+ (unsigned long long)addr, (unsigned long long)size);
+ }
+ }
+
+ depth = nextoffset = 0;
+ shift = 4; /* 4 spaces per indent */
+ do {
+ const struct fdt_property *prop;
+ const char *name;
+ int show;
+ int len;
+
+ offset = nextoffset;
+
+ /*
+ * Work out the file offset of this offset, and decide
+ * whether it is in the region list or not
+ */
+ file_ofs = base + offset;
+ if (reg < reg_end && file_ofs >= reg->offset + reg->size)
+ reg++;
+ in_region = reg < reg_end && file_ofs >= reg->offset &&
+ file_ofs < reg->offset + reg->size;
+ tag = fdt_next_tag(blob, offset, &nextoffset);
+
+ if (tag == FDT_END)
+ break;
+ show = in_region || disp->all;
+ if (show && disp->diff)
+ fprintf(f, "%c", in_region ? '+' : '-');
+
+ if (!show) {
+ /* Do this here to avoid 'if (show)' in every 'case' */
+ if (tag == FDT_BEGIN_NODE)
+ depth++;
+ else if (tag == FDT_END_NODE)
+ depth--;
+ continue;
+ }
+ if (tag != FDT_END) {
+ if (disp->show_addr)
+ fprintf(f, "%4x: ", file_ofs);
+ if (disp->show_offset)
+ fprintf(f, "%4x: ", file_ofs - base);
+ }
+
+ /* Green means included, red means excluded */
+ if (disp->colour)
+ print_ansi_colour(f, in_region ? COL_GREEN : COL_RED);
+
+ switch (tag) {
+ case FDT_PROP:
+ prop = fdt_get_property_by_offset(blob, offset, NULL);
+ name = fdt_string(blob, fdt32_to_cpu(prop->nameoff));
+ fprintf(f, "%*s%s", depth * shift, "", name);
+ utilfdt_print_data(prop->data,
+ fdt32_to_cpu(prop->len));
+ fprintf(f, ";");
+ break;
+
+ case FDT_NOP:
+ fprintf(f, "%*s// [NOP]", depth * shift, "");
+ break;
+
+ case FDT_BEGIN_NODE:
+ name = fdt_get_name(blob, offset, &len);
+ fprintf(f, "%*s%s {", depth++ * shift, "",
+ *name ? name : "/");
+ break;
+
+ case FDT_END_NODE:
+ fprintf(f, "%*s};", --depth * shift, "");
+ break;
+ }
+
+ /* Reset colour back to normal before end of line */
+ if (disp->colour)
+ print_ansi_colour(f, COL_NONE);
+ fprintf(f, "\n");
+ } while (1);
+
+ /* Print a list of strings if requested */
+ if (disp->list_strings) {
+ const char *str;
+ int offset;
+ int str_base = fdt_off_dt_strings(blob);
+
+ for (offset = 0; offset < fdt_size_dt_strings(blob);
+ offset += strlen(str) + 1) {
+ str = fdt_string(blob, offset);
+ int len = strlen(str) + 1;
+ int show;
+
+ /* Only print strings that are in the region */
+ file_ofs = str_base + offset;
+ in_region = reg < reg_end &&
+ file_ofs >= reg->offset &&
+ file_ofs + len < reg->offset +
+ reg->size;
+ show = in_region || disp->all;
+ if (show && disp->diff)
+ printf("%c", in_region ? '+' : '-');
+ if (disp->show_addr)
+ printf("%4x: ", file_ofs);
+ if (disp->show_offset)
+ printf("%4x: ", offset);
+ printf("%s\n", str);
+ }
+ }
+
+ return 0;
+}
+
+/**
+ * dump_fdt_regions() - Dump regions of an FDT as binary data
+ *
+ * This dumps an FDT as binary, but only certain regions of it. This is the
+ * final stage of the grep - we have a list of regions we want to dump,
+ * and this function dumps them.
+ *
+ * The output of this function may or may not be a valid FDT. To ensure it
+ * is, these disp->flags must be set:
+ *
+ * FDT_REG_SUPERNODES: ensures that subnodes are preceeded by their
+ * parents. Without this option, fragments of subnode data may be
+ * output without the supernodes above them. This is useful for
+ * hashing but cannot produce a valid FDT.
+ * FDT_REG_ADD_STRING_TAB: Adds a string table to the end of the FDT.
+ * Without this none of the properties will have names
+ * FDT_REG_ADD_MEM_RSVMAP: Adds a mem_rsvmap table - an FDT is invalid
+ * without this.
+ *
+ * @disp: Display structure, holding info about our options
+ * @blob: FDT blob to display
+ * @region: List of regions to display
+ * @count: Number of regions
+ */
+static int dump_fdt_regions(struct display_info *disp, const void *blob,
+ struct fdt_region region[], int count)
+{
+ struct fdt_header hdr, *fdt = &hdr;
+ int size, struct_start;
+ FILE *f = disp->fout;
+ int i;
+
+ /* Set up a basic header (even if we don't actually write it) */
+ memset(fdt, '\0', sizeof(*fdt));
+ fdt_set_magic(fdt, FDT_MAGIC);
+ struct_start = FDT_ALIGN(sizeof(struct fdt_header),
+ sizeof(struct fdt_reserve_entry));
+ fdt_set_off_mem_rsvmap(fdt, struct_start);
+ fdt_set_version(fdt, FDT_LAST_SUPPORTED_VERSION);
+ fdt_set_last_comp_version(fdt, FDT_FIRST_SUPPORTED_VERSION);
+
+ /*
+ * Calculate the total size of the regions we are writing out. The
+ * first will be the mem_rsvmap if the FDT_REG_ADD_MEM_RSVMAP flag
+ * is set. The last will be the string table if FDT_REG_ADD_STRING_TAB
+ * is set.
+ */
+ for (i = size = 0; i < count; i++)
+ size += region[i].size;
+
+ /* Bring in the mem_rsvmap section from the old file if requested */
+ if (count > 0 && (disp->flags & FDT_REG_ADD_MEM_RSVMAP)) {
+ struct_start += region[0].size;
+ size -= region[0].size;
+ }
+ fdt_set_off_dt_struct(fdt, struct_start);
+
+ /* Update the header to have the correct offsets/sizes */
+ if (count >= 2 && (disp->flags & FDT_REG_ADD_STRING_TAB)) {
+ int str_size;
+
+ str_size = region[count - 1].size;
+ fdt_set_size_dt_struct(fdt, size - str_size);
+ fdt_set_off_dt_strings(fdt, struct_start + size - str_size);
+ fdt_set_size_dt_strings(fdt, str_size);
+ fdt_set_totalsize(fdt, struct_start + size);
+ }
+
+ /* Write the header if required */
+ if (disp->header) {
+ if (sizeof(hdr) != fwrite(fdt, 1, sizeof(hdr), f))
+ return -1;
+ for (i = sizeof(hdr); i < fdt_off_mem_rsvmap(&hdr); i++)
+ fputc('\0', f);
+ }
+
+ /* Output all the nodes including any mem_rsvmap/string table */
+ for (i = size = 0; i < count; i++) {
+ struct fdt_region *reg = ®ion[i];
+
+ if (reg->size != fwrite((const char *)blob + reg->offset, 1,
+ reg->size, f))
+ return -1;
+ size += reg->size;
+ }
+
+ return 0;
+}
+
+/**
+ * show_region_list() - Print out a list of regions
+ *
+ * The list includes the region offset (absolute offset from start of FDT
+ * blob in bytes) and size
+ *
+ * @reg: List of regions to print
+ * @count: Number of regions
+ */
+static void show_region_list(struct fdt_region *reg, int count)
+{
+ int i;
+
+ printf("Regions: %d\n", count);
+ for (i = 0; i < count; i++, reg++) {
+ printf("%d: %-10x %-10x\n", i, reg->offset,
+ reg->offset + reg->size);
+ }
+}
+
+static int check_type_include(void *priv, int type, const char *data, int size)
+{
+ struct display_info *disp = priv;
+ struct value_node *val;
+ int match, none_match = FDT_IS_ANY;
+
+ /* If none of our conditions mention this type, we know nothing */
+ debug("type=%x, data=%s\n", type, data);
+ if (!((disp->types_inc | disp->types_exc) & type)) {
+ debug(" - not in any condition\n");
+ return -1;
+ }
+
+ /*
+ * Go through the list of conditions. For inclusive conditions, we
+ * return 1 at the first match. For exclusive conditions, we must
+ * check that there are no matches.
+ */
+ for (val = disp->value_head; val; val = val->next) {
+ if (!(type & val->type))
+ continue;
+ match = fdt_stringlist_contains(data, size, val->string);
+ debug(" - val->type=%x, str='%s', match=%d\n",
+ val->type, val->string, match);
+ if (match && val->include) {
+ debug(" - match inc %s\n", val->string);
+ return 1;
+ }
+ if (match)
+ none_match &= ~val->type;
+ }
+
+ /*
+ * If this is an exclusive condition, and nothing matches, then we
+ * should return 1.
+ */
+ if ((type & disp->types_exc) && (none_match & type)) {
+ debug(" - match exc\n");
+ /*
+ * Allow FDT_IS_COMPAT to make the final decision in the
+ * case where there is no specific type
+ */
+ if (type == FDT_IS_NODE && disp->types_exc == FDT_IS_ANY) {
+ debug(" - supressed exc node\n");
+ return -1;
+ }
+ return 1;
+ }
+
+ /*
+ * Allow FDT_IS_COMPAT to make the final decision in the
+ * case where there is no specific type (inclusive)
+ */
+ if (type == FDT_IS_NODE && disp->types_inc == FDT_IS_ANY)
+ return -1;
+
+ debug(" - no match, types_inc=%x, types_exc=%x, none_match=%x\n",
+ disp->types_inc, disp->types_exc, none_match);
+
+ return 0;
+}
+
+/**
+ * h_include() - Include handler function for fdt_find_regions()
+ *
+ * This function decides whether to include or exclude a node, property or
+ * compatible string. The function is defined by fdt_find_regions().
+ *
+ * The algorithm is documented in the code - disp->invert is 0 for normal
+ * operation, and 1 to invert the sense of all matches.
+ *
+ * See
+ */
+static int h_include(void *priv, const void *fdt, int offset, int type,
+ const char *data, int size)
+{
+ struct display_info *disp = priv;
+ int inc, len;
+
+ inc = check_type_include(priv, type, data, size);
+
+ /*
+ * If the node name does not tell us anything, check the
+ * compatible string
+ */
+ if (inc == -1 && type == FDT_IS_NODE) {
+ debug(" - checking compatible2\n");
+ data = fdt_getprop(fdt, offset, "compatible", &len);
+ inc = check_type_include(priv, FDT_IS_COMPAT, data, len);
+ }
+
+ switch (inc) {
+ case 1:
+ inc = !disp->invert;
+ break;
+ case 0:
+ inc = disp->invert;
+ break;
+ }
+ debug(" - returning %d\n", inc);
+
+ return inc;
+}
+
+static int fdt_find_regions(const void *fdt,
+ int (*h_include)(void *priv, const void *fdt, int offset,
+ int type, const char *data, int size),
+ struct display_info *disp, struct fdt_region *region,
+ int max_regions, char *path, int path_len, int flags)
+{
+ struct fdt_region_state state;
+ int count;
+ int ret;
+
+ count = 0;
+ ret = fdt_first_region(fdt, h_include, disp,
+ ®ion[count++], path, path_len,
+ disp->flags, &state);
+ while (ret == 0) {
+ ret = fdt_next_region(fdt, h_include, disp,
+ ®ion[count], path, path_len,
+ disp->flags, &state);
+ if (!ret)
+ count++;
+ }
+
+ if (ret != -FDT_ERR_NOTFOUND)
+ return ret;
+
+ return count;
+}
+
+/**
+ * Run the main fdtgrep operation, given a filename and valid arguments
+ *
+ * @param disp Display information / options
+ * @param filename Filename of blob file
+ * @param return 0 if ok, -ve on error
+ */
+static int do_fdtgrep(struct display_info *disp, const char *filename)
+{
+ struct fdt_region *region;
+ int max_regions;
+ int count = 100;
+ char path[1024];
+ char *blob;
+ int i, ret;
+
+ blob = utilfdt_read(filename);
+ if (!blob)
+ return -1;
+ ret = fdt_check_header(blob);
+ if (ret) {
+ fprintf(stderr, "Error: %s\n", fdt_strerror(ret));
+ return ret;
+ }
+
+ /* Allow old files, but they are untested */
+ if (fdt_version(blob) < 17 && disp->value_head) {
+ fprintf(stderr, "Warning: fdtgrep does not fully support"
+ " version %d files\n", fdt_version(blob));
+ }
+
+ /*
+ * We do two passes, since we don't know how many regions we need.
+ * The first pass will count the regions, but if it is too many,
+ * we do another pass to actually record them.
+ */
+ for (i = 0; i < 2; i++) {
+ region = malloc(count * sizeof(struct fdt_region));
+ if (!region) {
+ fprintf(stderr, "Out of memory for %d regions\n",
+ count);
+ return -1;
+ }
+ max_regions = count;
+ count = fdt_find_regions(blob,
+ h_include, disp,
+ region, max_regions, path, sizeof(path),
+ disp->flags);
+ if (count < 0) {
+ report_error("fdt_find_regions", count);
+ return -1;
+ }
+ if (count <= max_regions)
+ break;
+ free(region);
+ }
+
+ /* Optionally print a list of regions */
+ if (disp->region_list)
+ show_region_list(region, count);
+
+ /* Output either source .dts or binary .dtb */
+ if (disp->output == OUT_DTS) {
+ ret = display_fdt_by_regions(disp, blob, region, count);
+ } else {
+ ret = dump_fdt_regions(disp, blob, region, count);
+ if (ret)
+ fprintf(stderr, "Write failure\n");
+ }
+
+ free(blob);
+ free(region);
+
+ return ret;
+}
+
+static const char usage_synopsis[] =
+ "fdtgrep - extract portions from device tree\n"
+ "\n"
+ "Usage:\n"
+ " fdtgrep <options> <dt file>|-\n\n"
+ "Output formats are:\n"
+ "\tdts - device tree soure text\n"
+ "\tdtb - device tree blob (sets -Hmt automatically)\n"
+ "\tbin - device tree fragment (may not be a valid .dtb)";
+
+static const char usage_short_opts[] = "haAc:C:defg:G:HIlLmn:N:o:O:p:P:sStv"
+ USAGE_COMMON_SHORT_OPTS;
+static struct option const usage_long_opts[] = {
+ {"show-address", no_argument, NULL, 'a'},
+ {"colour", no_argument, NULL, 'A'},
+ {"include-compat", a_argument, NULL, 'c'},
+ {"exclude-compat", a_argument, NULL, 'C'},
+ {"diff", no_argument, NULL, 'd'},
+ {"enter-node", no_argument, NULL, 'e'},
+ {"show-offset", no_argument, NULL, 'f'},
+ {"include-match", a_argument, NULL, 'g'},
+ {"exclude-match", a_argument, NULL, 'G'},
+ {"show-header", no_argument, NULL, 'H'},
+ {"show-version", no_argument, NULL, 'I'},
+ {"list-regions", no_argument, NULL, 'l'},
+ {"list-strings", no_argument, NULL, 'L'},
+ {"include-mem", no_argument, NULL, 'm'},
+ {"include-node", a_argument, NULL, 'n'},
+ {"exclude-node", a_argument, NULL, 'N'},
+ {"include-prop", a_argument, NULL, 'p'},
+ {"exclude-prop", a_argument, NULL, 'P'},
+ {"show-subnodes", no_argument, NULL, 's'},
+ {"skip-supernodes", no_argument, NULL, 'S'},
+ {"show-stringtab", no_argument, NULL, 't'},
+ {"out", a_argument, NULL, 'o'},
+ {"out-format", a_argument, NULL, 'O'},
+ {"invert-match", no_argument, NULL, 'v'},
+ USAGE_COMMON_LONG_OPTS,
+};
+static const char * const usage_opts_help[] = {
+ "Display address",
+ "Show all nodes/tags, colour those that match",
+ "Compatible nodes to include in grep",
+ "Compatible nodes to exclude in grep",
+ "Diff: Mark matching nodes with +, others with -",
+ "Enter direct subnode names of matching nodes",
+ "Display offset",
+ "Node/property/compatible string to include in grep",
+ "Node/property/compatible string to exclude in grep",
+ "Output a header",
+ "Put \"/dts-v1/;\" on first line of dts output",
+ "Output a region list",
+ "List strings in string table",
+ "Include mem_rsvmap section in binary output",
+ "Node to include in grep",
+ "Node to exclude in grep",
+ "Property to include in grep",
+ "Property to exclude in grep",
+ "Show all subnodes matching nodes",
+ "Don't include supernodes of matching nodes",
+ "Include string table in binary output",
+ "-o <output file>",
+ "-O <output format>",
+ "Invert the sense of matching (select non-matching lines)",
+ USAGE_COMMON_OPTS_HELP
+};
+
+static void scan_args(struct display_info *disp, int argc, char *argv[])
+{
+ int opt;
+
+ while ((opt = util_getopt_long()) != EOF) {
+ int type = 0;
+ int inc = 1;
+
+ switch (opt) {
+ case_USAGE_COMMON_FLAGS
+ case 'a':
+ disp->show_addr = 1;
+ break;
+ case 'A':
+ disp->all = 1;
+ break;
+ case 'C':
+ inc = 0;
+ /* no break */
+ case 'c':
+ type = FDT_IS_COMPAT;
+ break;
+ case 'd':
+ disp->diff = 1;
+ break;
+ case 'e':
+ disp->flags |= FDT_REG_DIRECT_SUBNODES;
+ break;
+ case 'f':
+ disp->show_offset = 1;
+ break;
+ case 'G':
+ inc = 0;
+ /* no break */
+ case 'g':
+ type = FDT_IS_ANY;
+ break;
+ case 'H':
+ disp->header = 1;
+ break;
+ case 'l':
+ disp->region_list = 1;
+ break;
+ case 'L':
+ disp->list_strings = 1;
+ break;
+ case 'm':
+ disp->flags |= FDT_REG_ADD_MEM_RSVMAP;
+ break;
+ case 'N':
+ inc = 0;
+ /* no break */
+ case 'n':
+ type = FDT_IS_NODE;
+ break;
+ case 'o':
+ disp->output_fname = optarg;
+ break;
+ case 'O':
+ if (!strcmp(optarg, "dtb"))
+ disp->output = OUT_DTB;
+ else if (!strcmp(optarg, "dts"))
+ disp->output = OUT_DTS;
+ else if (!strcmp(optarg, "bin"))
+ disp->output = OUT_BIN;
+ else
+ usage("Unknown output format");
+ break;
+ case 'P':
+ inc = 0;
+ /* no break */
+ case 'p':
+ type = FDT_IS_PROP;
+ break;
+ case 's':
+ disp->flags |= FDT_REG_ALL_SUBNODES;
+ break;
+ case 'S':
+ disp->flags &= ~FDT_REG_SUPERNODES;
+ break;
+ case 't':
+ disp->flags |= FDT_REG_ADD_STRING_TAB;
+ break;
+ case 'v':
+ disp->invert = 1;
+ break;
+ case 'I':
+ disp->show_dts_version = 1;
+ break;
+ }
+
+ if (type && value_add(disp, &disp->value_head, type, inc,
+ optarg))
+ usage("Cannot add value");
+ }
+
+ if (disp->invert && disp->types_exc)
+ usage("-v has no meaning when used with 'exclude' conditions");
+}
+
+int main(int argc, char *argv[])
+{
+ char *filename = NULL;
+ struct display_info disp;
+ int ret;
+
+ /* set defaults */
+ memset(&disp, '\0', sizeof(disp));
+ disp.flags = FDT_REG_SUPERNODES; /* Default flags */
+
+ scan_args(&disp, argc, argv);
+
+ /* Show matched lines in colour if we can */
+ disp.colour = disp.all && isatty(0);
+
+ /* Any additional arguments can match anything, just like -g */
+ while (optind < argc - 1) {
+ if (value_add(&disp, &disp.value_head, FDT_IS_ANY, 1,
+ argv[optind++]))
+ usage("Cannot add value");
+ }
+
+ if (optind < argc)
+ filename = argv[optind++];
+ if (!filename)
+ usage("Missing filename");
+
+ /* If a valid .dtb is required, set flags to ensure we get one */
+ if (disp.output == OUT_DTB) {
+ disp.header = 1;
+ disp.flags |= FDT_REG_ADD_MEM_RSVMAP | FDT_REG_ADD_STRING_TAB;
+ }
+
+ if (disp.output_fname) {
+ disp.fout = fopen(disp.output_fname, "w");
+ if (!disp.fout)
+ usage("Cannot open output file");
+ } else {
+ disp.fout = stdout;
+ }
+
+ /* Run the grep and output the results */
+ ret = do_fdtgrep(&disp, filename);
+ if (disp.output_fname)
+ fclose(disp.fout);
+ if (ret)
+ return 1;
+
+ return 0;
+}
diff --git a/tests/grep.dts b/tests/grep.dts
new file mode 100644
index 0000000..9600657
--- /dev/null
+++ b/tests/grep.dts
@@ -0,0 +1,23 @@
+/dts-v1/;
+/memreserve/ 1 2;
+/ {
+ model = "MyBoardName";
+ compatible = "MyBoardName", "MyBoardFamilyName";
+ #address-cells = <2>;
+ #size-cells = <2>;
+ chosen {
+ bootargs = "root=/dev/sda2";
+ linux,platform = <0x600>;
+ };
+ holiday {
+ compatible = "ixtapa", "mexico";
+ weather = "sunny";
+ status = "okay";
+ flight@1 {
+ airline = "alaska";
+ };
+ flight@2 {
+ airline = "lan";
+ };
+ };
+};
diff --git a/tests/run_tests.sh b/tests/run_tests.sh
index 8757c11..31b5148 100755
--- a/tests/run_tests.sh
+++ b/tests/run_tests.sh
@@ -592,6 +592,357 @@ utilfdt_tests () {
run_test utilfdt_test
}
+# Add a property to a tree, then hash it and see if it changed
+# Args:
+# $1: 0 if we expect it to stay the same, 1 if we expect a change
+# $2: node to add a property to
+# $3: arguments for fdtget
+# $4: filename of device tree binary
+# $5: hash of unchanged file (empty to calculate it)
+# $6: node to add a property to ("testing" by default if empty)
+check_hash () {
+ local changed="$1"
+ local node="$2"
+ local args="$3"
+ local tree="$4"
+ local base="$5"
+ local nodename="$6"
+
+ if [ -z "$nodename" ]; then
+ nodename=testing
+ fi
+ if [ -z "$base" ]; then
+ base=$($DTGREP ${args} -O bin $tree | sha1sum)
+ fi
+ $DTPUT $tree $node $nodename 1
+ hash=$($DTGREP ${args} -O bin $tree | sha1sum)
+ if [ "$base" == "$hash" ]; then
+ if [ "$changed" == 1 ]; then
+ echo "$test: Hash should have changed"
+ echo base $base
+ echo hash $hash
+ false
+ fi
+ else
+ if [ "$changed" == 0 ]; then
+ echo "$test: Base hash is $base but it was changed to $hash"
+ false
+ fi
+ fi
+}
+
+# Check the number of lines generated matches what we expect
+# Args:
+# $1: Expected number of lines
+# $2...: Command line to run to generate output
+check_lines () {
+ local base="$1"
+
+ shift
+ lines=$($@ | wc -l)
+ if [ "$base" != "$lines" ]; then
+ echo "Expected $base lines but got $lines lines"
+ false
+ fi
+}
+
+# Check the number of bytes generated matches what we expect
+# Args:
+# $1: Expected number of bytes
+# $2...: Command line to run to generate output
+check_bytes () {
+ local base="$1"
+
+ shift
+ bytes=$($@ | wc -c)
+ if [ "$base" != "$bytes" ]; then
+ echo "Expected $base bytes but got $bytes bytes"
+ false
+ fi
+}
+
+# Check whether a command generates output which contains a string
+# Args:
+# $1: 0 to expect the string to be absent, 1 to expect it to be present
+# $2: text to grep for
+# $3...: Command to execute
+check_contains () {
+ contains="$1"
+ text="$2"
+
+ shift 2
+ if $@ | grep -q $text; then
+ if [ $contains -ne 1 ]; then
+ echo "Did not expect to find $text in output"
+ false
+ fi
+ else
+ if [ $contains -ne 0 ]; then
+ echo "Expected to find $text in output"
+ false
+ fi
+ fi
+}
+
+# Check that $2 and $3 are equal. $1 is the test name to display
+equal_test () {
+ echo -n "$1: "
+ if [ "$2" == "$3" ]; then
+ PASS
+ else
+ FAIL "$2 != $3"
+ fi
+}
+
+fdtgrep_tests () {
+ local addr
+ local all_lines # Total source lines in .dts file
+ local base
+ local dt_start
+ local lines
+ local node_lines # Number of lines of 'struct' output
+ local orig
+ local string_size
+ local tmp
+ local tree
+
+ tmp=/tmp/tests.$$
+ orig=region_tree.test.dtb
+ run_wrap_test ./region_tree 0 1000 ${orig}
+
+ # Hash of partial tree
+ # - modify tree in various ways and check that hash is unaffected
+ tree=region_tree.mod.dtb
+ cp $orig $tree
+ args="-n /images/kernel@1"
+ run_wrap_test check_hash 0 /images "$args" $tree
+ run_wrap_test check_hash 0 /images/kernel@1/hash@1 "$args" $tree
+ run_wrap_test check_hash 0 / "$args" $tree
+ $DTPUT -c $tree /images/kernel@1/newnode
+ run_wrap_test check_hash 0 / "$args" $tree
+ run_wrap_test check_hash 1 /images/kernel@1 "$args" $tree
+
+ # Now hash immediate subnodes so we detect a new subnode added
+ cp $orig $tree
+ args="-n /images/kernel@1 -e"
+ run_wrap_test check_hash 0 /images "$args" $tree
+ run_wrap_test check_hash 0 /images/kernel@1/hash@1 "$args" $tree
+ run_wrap_test check_hash 0 / "$args" $tree
+ base=$($DTGREP $args -O bin $tree | sha1sum)
+ $DTPUT -c $tree /images/kernel@1/newnode
+ run_wrap_test check_hash 1 / "$args" $tree "$base"
+ cp $orig $tree
+ run_wrap_test check_hash 1 /images/kernel@1 "$args" $tree
+
+ # Hash the string table, which should change if we add a new property name
+ # (Adding an existing property name will just reuse that string)
+ cp $orig $tree
+ args="-t -n /images/kernel@1"
+ run_wrap_test check_hash 0 /images "$args" $tree "" data
+ run_wrap_test check_hash 1 /images/kernel@1 "$args" $tree
+
+ dts=grep.dts
+ dtb=grep.dtb
+ run_dtc_test -O dtb -p 0x1000 -o $dtb $dts
+
+ # Tests for each argument are roughly in alphabetical order
+ #
+ # First a sanity check that we can get back the source from the .dtb
+ all_lines=$(cat $dts | wc -l)
+ run_wrap_test check_lines ${all_lines} $DTGREP -Im $dtb
+ node_lines=$(($all_lines - 2))
+
+ # Get the offset of the dt_struct start (also tests -H somewhat)
+ dt_start=$($DTGREP -H $dtb | awk '/off_dt_struct:/ {print $3}')
+ dt_size=$($DTGREP -H $dtb | awk '/size_dt_struct:/ {print $3}')
+
+ # Check -a: the first line should contain the offset of the dt_start
+ addr=$($DTGREP -a $dtb | head -1 | tr -d : | awk '{print $1}')
+ run_wrap_test equal_test "-a offset first" "$dt_start" "0x$addr"
+
+ # Last line should be 8 bytes less than the size (NODE, END tags)
+ addr=$($DTGREP -a $dtb | tail -1 | tr -d : | awk '{print $1}')
+ last=$(printf "%#x" $(($dt_start + $dt_size - 8)))
+ run_wrap_test equal_test "-a offset last" "$last" "0x$addr"
+
+ # Check the offset option in a similar way. The first offset should be 0
+ # and the last one should be the size of the struct area.
+ addr=$($DTGREP -f $dtb | head -1 | tr -d : | awk '{print $1}')
+ run_wrap_test equal_test "-o offset first" "0x0" "0x$addr"
+ addr=$($DTGREP -f $dtb | tail -1 | tr -d : | awk '{print $1}')
+ last=$(printf "%#x" $(($dt_size - 8)))
+ run_wrap_test equal_test "-f offset last" "$last" "0x$addr"
+
+ # Check that -A controls display of all lines
+ # The 'chosen' node should only have four output lines
+ run_wrap_test check_lines $node_lines $DTGREP -S -A -n /chosen $dtb
+ run_wrap_test check_lines 4 $DTGREP -S -n /chosen $dtb
+
+ # Check that -c picks out nodes
+ run_wrap_test check_lines 5 $DTGREP -S -c ixtapa $dtb
+ run_wrap_test check_lines $(($node_lines - 5)) $DTGREP -S -C ixtapa $dtb
+
+ # -d marks selected lines with +
+ run_wrap_test check_lines $node_lines $DTGREP -S -Ad -n /chosen $dtb
+ run_wrap_test check_lines 4 $DTGREP -S -Ad -n /chosen $dtb |grep +
+
+ # -g should find a node, property or compatible string
+ run_wrap_test check_lines 2 $DTGREP -S -g / $dtb
+ run_wrap_test check_lines 2 $DTGREP -S -g /chosen $dtb
+ run_wrap_test check_lines $(($node_lines - 2)) $DTGREP -S -G /chosen $dtb
+
+ run_wrap_test check_lines 1 $DTGREP -S -g bootargs $dtb
+ run_wrap_test check_lines $(($node_lines - 1)) $DTGREP -S -G bootargs $dtb
+
+ # We should find the /holiday node, so 1 line for 'holiday {', one for '}'
+ run_wrap_test check_lines 2 $DTGREP -S -g ixtapa $dtb
+ run_wrap_test check_lines $(($node_lines - 2)) $DTGREP -S -G ixtapa $dtb
+
+ run_wrap_test check_lines 3 $DTGREP -S -g ixtapa -g bootargs $dtb
+ run_wrap_test check_lines $(($node_lines - 3)) $DTGREP -S -G ixtapa \
+ -G bootargs $dtb
+
+ # -l outputs a,list of regions - here we should get 3: one for the header,
+ # one for the node and one for the 'end' tag.
+ run_wrap_test check_lines 3 $DTGREP -S -l -n /chosen $dtb -o $tmp
+
+ # -L outputs all the strings in the string table
+ cat >$tmp <<END
+ #address-cells
+ airline
+ bootargs
+ compatible
+ linux,platform
+ model
+ #size-cells
+ status
+ weather
+END
+ lines=$(cat $tmp | wc -l)
+ run_wrap_test check_lines $lines $DTGREP -S -L -n // $dtb
+
+ # Check that the -m flag works
+ run_wrap_test check_contains 1 memreserve $DTGREP -Im $dtb
+ run_wrap_test check_contains 0 memreserve $DTGREP -I $dtb
+
+ # Test -n
+ run_wrap_test check_lines 0 $DTGREP -S -n // $dtb
+ run_wrap_test check_lines 0 $DTGREP -S -n chosen $dtb
+ run_wrap_test check_lines 0 $DTGREP -S -n holiday $dtb
+ run_wrap_test check_lines 0 $DTGREP -S -n \"\" $dtb
+ run_wrap_test check_lines 4 $DTGREP -S -n /chosen $dtb
+ run_wrap_test check_lines 5 $DTGREP -S -n /holiday $dtb
+ run_wrap_test check_lines 9 $DTGREP -S -n /chosen -n /holiday $dtb
+
+ # Test -N which should list everything except matching nodes
+ run_wrap_test check_lines $node_lines $DTGREP -S -N // $dtb
+ run_wrap_test check_lines $node_lines $DTGREP -S -N chosen $dtb
+ run_wrap_test check_lines $(($node_lines - 4)) $DTGREP -S -N /chosen $dtb
+ run_wrap_test check_lines $(($node_lines - 5)) $DTGREP -S -N /holiday $dtb
+ run_wrap_test check_lines $(($node_lines - 9)) $DTGREP -S -N /chosen \
+ -N /holiday $dtb
+
+ # Using -n and -N together is undefined, so we don't have tests for that
+ # The same applies for -p/-P and -c/-C.
+ run_wrap_error_test $DTGREP -n chosen -N holiday $dtb
+ run_wrap_error_test $DTGREP -c chosen -C holiday $dtb
+ run_wrap_error_test $DTGREP -p chosen -P holiday $dtb
+
+ # Test -o: this should output just the .dts file to a file
+ # Where there is non-dts output it should go to stdout
+ rm -f $tmp
+ run_wrap_test check_lines 0 $DTGREP $dtb -o $tmp
+ run_wrap_test check_lines $node_lines cat $tmp
+
+ # Here we expect a region list with a single entry, plus a header line
+ # on stdout
+ run_wrap_test check_lines 2 $DTGREP $dtb -o $tmp -l
+ run_wrap_test check_lines $node_lines cat $tmp
+
+ # Here we expect a list of strings on stdout
+ run_wrap_test check_lines ${lines} $DTGREP $dtb -o $tmp -L
+ run_wrap_test check_lines $node_lines cat $tmp
+
+ # Test -p: with -S we only get the compatible lines themselves
+ run_wrap_test check_lines 2 $DTGREP -S -p compatible -n // $dtb
+ run_wrap_test check_lines 1 $DTGREP -S -p bootargs -n // $dtb
+
+ # Without -S we also get the node containing these properties
+ run_wrap_test check_lines 6 $DTGREP -p compatible -n // $dtb
+ run_wrap_test check_lines 5 $DTGREP -p bootargs -n // $dtb
+
+ # Now similar tests for -P
+ # First get the number of property lines (containing '=')
+ lines=$(grep "=" $dts |wc -l)
+ run_wrap_test check_lines $(($lines - 2)) $DTGREP -S -P compatible \
+ -n // $dtb
+ run_wrap_test check_lines $(($lines - 1)) $DTGREP -S -P bootargs \
+ -n // $dtb
+ run_wrap_test check_lines $(($lines - 3)) $DTGREP -S -P compatible \
+ -P bootargs -n // $dtb
+
+ # Without -S we also get the node containing these properties
+ run_wrap_test check_lines $(($node_lines - 2)) $DTGREP -P compatible \
+ -n // $dtb
+ run_wrap_test check_lines $(($node_lines - 1)) $DTGREP -P bootargs \
+ -n // $dtb
+ run_wrap_test check_lines $(($node_lines - 3)) $DTGREP -P compatible \
+ -P bootargs -n // $dtb
+
+ # -s should bring in all sub-nodes
+ run_wrap_test check_lines 2 $DTGREP -p none -n / $dtb
+ run_wrap_test check_lines 6 $DTGREP -e -p none -n / $dtb
+ run_wrap_test check_lines 2 $DTGREP -S -p none -n /holiday $dtb
+ run_wrap_test check_lines 4 $DTGREP -p none -n /holiday $dtb
+ run_wrap_test check_lines 8 $DTGREP -e -p none -n /holiday $dtb
+
+ # -v inverts the polarity of any condition
+ run_wrap_test check_lines $(($node_lines - 2)) $DTGREP -Sv -p none \
+ -n / $dtb
+ run_wrap_test check_lines $(($node_lines - 2)) $DTGREP -Sv -p compatible \
+ -n // $dtb
+ run_wrap_test check_lines $(($node_lines - 2)) $DTGREP -Sv -g /chosen \
+ $dtb
+ run_wrap_test check_lines $node_lines $DTGREP -Sv -n // $dtb
+ run_wrap_test check_lines $node_lines $DTGREP -Sv -n chosen $dtb
+ run_wrap_error_test $DTGREP -v -N holiday $dtb
+
+ # Check that the -I flag works
+ run_wrap_test check_contains 1 dts-v1 $DTGREP -I $dtb
+ run_wrap_test check_contains 0 dts-v1 $DTGREP $dtb
+
+ # Now some dtb tests. The dts tests above have tested the basic grepping
+ # features so we only need to concern ourselves with things that are
+ # different about dtb/bin output.
+
+ # An empty node list should just give us the FDT_END tag
+ run_wrap_test check_bytes 4 $DTGREP -n // -O bin $dtb
+
+ # The mem_rsvmap is two entries of 16 bytes each
+ run_wrap_test check_bytes $((4 + 32)) $DTGREP -m -n // -O bin $dtb
+
+ # Check we can add the string table
+ string_size=$($DTGREP -H $dtb | awk '/size_dt_strings:/ {print $3}')
+ run_wrap_test check_bytes $((4 + $string_size)) $DTGREP -t -n // -O bin \
+ $dtb
+ run_wrap_test check_bytes $((4 + 32 + $string_size)) $DTGREP -tm \
+ -n // -O bin $dtb
+
+ # Check that a pass-through works ok. fdtgrep aligns the mem_rsvmap table
+ # to a 16-bytes boundary, but dtc uses 8 bytes so we expect the size to
+ # increase by 8 bytes...
+ run_dtc_test -O dtb -o $dtb $dts
+ base=$(stat -c %s $dtb)
+ run_wrap_test check_bytes $(($base + 8)) $DTGREP -O dtb $dtb
+
+ # ...but we should get the same output from fdtgrep in a second pass
+ run_wrap_test check_bytes 0 $DTGREP -O dtb $dtb -o $tmp
+ base=$(stat -c %s $tmp)
+ run_wrap_test check_bytes $base $DTGREP -O dtb $tmp
+
+ rm -f $tmp
+}
+
while getopts "vt:m" ARG ; do
case $ARG in
"v")
@@ -607,7 +958,7 @@ while getopts "vt:m" ARG ; do
done
if [ -z "$TESTSETS" ]; then
- TESTSETS="libfdt utilfdt dtc dtbs_equal fdtget fdtput"
+ TESTSETS="libfdt utilfdt dtc dtbs_equal fdtget fdtput fdtgrep"
fi
# Make sure we don't have stale blobs lying around
@@ -633,6 +984,9 @@ for set in $TESTSETS; do
"fdtput")
fdtput_tests
;;
+ "fdtgrep")
+ fdtgrep_tests
+ ;;
esac
done
diff --git a/tests/tests.sh b/tests/tests.sh
index 31530d5..8e4b65b 100644
--- a/tests/tests.sh
+++ b/tests/tests.sh
@@ -21,6 +21,7 @@ FAIL_IF_SIGNAL () {
DTC=../dtc
DTGET=../fdtget
DTPUT=../fdtput
+DTGREP=../fdtgrep
verbose_run () {
if [ -z "$QUIET_TEST" ]; then
--
1.9.1.423.g4596e3a
--
To unsubscribe from this list: send the line "unsubscribe devicetree-compiler" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related [flat|nested] 9+ messages in thread
* Re: [RESEND PATCH v3 1/3] libfdt: Add function to find regions in an FDT
[not found] ` <1395500836-12735-2-git-send-email-sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
@ 2014-05-01 16:13 ` Simon Glass
[not found] ` <CAPnjgZ0ehgSd1WprpMd1ZY=LaZaKkBG3MaNPV5ja27Q2p12WvA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
0 siblings, 1 reply; 9+ messages in thread
From: Simon Glass @ 2014-05-01 16:13 UTC (permalink / raw)
To: Devicetree Compiler; +Cc: Jon Loeliger, David Gibson, Simon Glass
Hi,
On 22 March 2014 08:07, Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org> wrote:
> Given a set of nodes and properties, find the regions of the device tree
> which describe those parts.
>
> A test is provided which builds a tree while tracking where the regions
> should be, then calls fdt_first/next_region() to make sure that it agrees.
>
> Further tests will come as part of fdtgrep.
>
> Signed-off-by: Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
Have I sent this to the right place? Any comments?
Regards,
Simon
--
To unsubscribe from this list: send the line "unsubscribe devicetree-compiler" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [RESEND PATCH v3 1/3] libfdt: Add function to find regions in an FDT
[not found] ` <CAPnjgZ0ehgSd1WprpMd1ZY=LaZaKkBG3MaNPV5ja27Q2p12WvA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
@ 2014-05-02 5:40 ` David Gibson
[not found] ` <20140502054014.GF6035-1s0os16eZneny3qCrzbmXA@public.gmane.org>
0 siblings, 1 reply; 9+ messages in thread
From: David Gibson @ 2014-05-02 5:40 UTC (permalink / raw)
To: Simon Glass; +Cc: Devicetree Compiler, Jon Loeliger
[-- Attachment #1: Type: text/plain, Size: 1392 bytes --]
On Thu, May 01, 2014 at 09:13:51AM -0700, Simon Glass wrote:
> Hi,
>
> On 22 March 2014 08:07, Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org> wrote:
> > Given a set of nodes and properties, find the regions of the device tree
> > which describe those parts.
> >
> > A test is provided which builds a tree while tracking where the regions
> > should be, then calls fdt_first/next_region() to make sure that it agrees.
> >
> > Further tests will come as part of fdtgrep.
> >
> > Signed-off-by: Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
>
> Have I sent this to the right place? Any comments?
Sorry. I know you've resent this several times, and I've been
procrastinating about it since forever.
Basically, I'm just not convinced. For all your efforts to explain
the rationale, it just seems like a really ad-hoc set of flags and
conditions that doesn't seem to form a coherent whole.
From the lack of other responses, I'm assuming there's not really
anyone else who sees it as a compelling feature either.
So, sorry. For the forseeable future, you'll need to leave this
implemented in your own code, using but outside of libfdt itself.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
[-- Attachment #2: Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [RESEND PATCH v3 1/3] libfdt: Add function to find regions in an FDT
[not found] ` <20140502054014.GF6035-1s0os16eZneny3qCrzbmXA@public.gmane.org>
@ 2014-06-03 3:12 ` Simon Glass
[not found] ` <CAPnjgZ2OHco3GjH_rPC+7Fu36QqBK2_sWvcaF1WOWFpBNkZTcQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
0 siblings, 1 reply; 9+ messages in thread
From: Simon Glass @ 2014-06-03 3:12 UTC (permalink / raw)
To: David Gibson; +Cc: Devicetree Compiler, Jon Loeliger
Hi David,
On 1 May 2014 23:40, David Gibson <david-xT8FGy+AXnRB3Ne2BGzF6laj5H9X9Tb+@public.gmane.org> wrote:
> On Thu, May 01, 2014 at 09:13:51AM -0700, Simon Glass wrote:
>> Hi,
>>
>> On 22 March 2014 08:07, Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org> wrote:
>> > Given a set of nodes and properties, find the regions of the device tree
>> > which describe those parts.
>> >
>> > A test is provided which builds a tree while tracking where the regions
>> > should be, then calls fdt_first/next_region() to make sure that it agrees.
>> >
>> > Further tests will come as part of fdtgrep.
>> >
>> > Signed-off-by: Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
>>
>> Have I sent this to the right place? Any comments?
>
> Sorry. I know you've resent this several times, and I've been
> procrastinating about it since forever.
>
> Basically, I'm just not convinced. For all your efforts to explain
> the rationale, it just seems like a really ad-hoc set of flags and
> conditions that doesn't seem to form a coherent whole.
Are you referring to fdtgrep or the new library function? I'd be happy
enough getting the library function to start with in if there isn't
much need for the grep utility.
>
> From the lack of other responses, I'm assuming there's not really
> anyone else who sees it as a compelling feature either.
Is there normally a lot of mailing list traffic for new libfdt
features? This is used for verified boot in U-Boot for example.
>
> So, sorry. For the forseeable future, you'll need to leave this
> implemented in your own code, using but outside of libfdt itself.
Regards,
Simon
--
To unsubscribe from this list: send the line "unsubscribe devicetree-compiler" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [RESEND PATCH v3 1/3] libfdt: Add function to find regions in an FDT
[not found] ` <CAPnjgZ2OHco3GjH_rPC+7Fu36QqBK2_sWvcaF1WOWFpBNkZTcQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
@ 2014-06-05 14:12 ` David Gibson
[not found] ` <20140605141236.GA25412-RXTfZT5YzpxwFLYp8hBm2A@public.gmane.org>
0 siblings, 1 reply; 9+ messages in thread
From: David Gibson @ 2014-06-05 14:12 UTC (permalink / raw)
To: Simon Glass; +Cc: Devicetree Compiler, Jon Loeliger
[-- Attachment #1: Type: text/plain, Size: 1900 bytes --]
On Mon, Jun 02, 2014 at 09:12:55PM -0600, Simon Glass wrote:
> Hi David,
>
> On 1 May 2014 23:40, David Gibson <david-xT8FGy+AXnRB3Ne2BGzF6laj5H9X9Tb+@public.gmane.org> wrote:
> > On Thu, May 01, 2014 at 09:13:51AM -0700, Simon Glass wrote:
> >> Hi,
> >>
> >> On 22 March 2014 08:07, Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org> wrote:
> >> > Given a set of nodes and properties, find the regions of the device tree
> >> > which describe those parts.
> >> >
> >> > A test is provided which builds a tree while tracking where the regions
> >> > should be, then calls fdt_first/next_region() to make sure that it agrees.
> >> >
> >> > Further tests will come as part of fdtgrep.
> >> >
> >> > Signed-off-by: Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
> >>
> >> Have I sent this to the right place? Any comments?
> >
> > Sorry. I know you've resent this several times, and I've been
> > procrastinating about it since forever.
> >
> > Basically, I'm just not convinced. For all your efforts to explain
> > the rationale, it just seems like a really ad-hoc set of flags and
> > conditions that doesn't seem to form a coherent whole.
>
> Are you referring to fdtgrep or the new library function? I'd be happy
> enough getting the library function to start with in if there isn't
> much need for the grep utility.
The library function primarily.
> > From the lack of other responses, I'm assuming there's not really
> > anyone else who sees it as a compelling feature either.
>
> Is there normally a lot of mailing list traffic for new libfdt
> features? This is used for verified boot in U-Boot for example.
Hm, that's true I guess.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
[-- Attachment #2: Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [RESEND PATCH v3 1/3] libfdt: Add function to find regions in an FDT
[not found] ` <20140605141236.GA25412-RXTfZT5YzpxwFLYp8hBm2A@public.gmane.org>
@ 2015-05-07 1:08 ` Simon Glass
0 siblings, 0 replies; 9+ messages in thread
From: Simon Glass @ 2015-05-07 1:08 UTC (permalink / raw)
To: David Gibson; +Cc: Devicetree Compiler, Jon Loeliger
Hi David,
On 5 June 2014 at 08:12, David Gibson <david-xT8FGy+AXnRB3Ne2BGzF6laj5H9X9Tb+@public.gmane.org> wrote:
>
> On Mon, Jun 02, 2014 at 09:12:55PM -0600, Simon Glass wrote:
> > Hi David,
> >
> > On 1 May 2014 23:40, David Gibson <david-xT8FGy+AXnRB3Ne2BGzF6laj5H9X9Tb+@public.gmane.org> wrote:
> > > On Thu, May 01, 2014 at 09:13:51AM -0700, Simon Glass wrote:
> > >> Hi,
> > >>
> > >> On 22 March 2014 08:07, Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org> wrote:
> > >> > Given a set of nodes and properties, find the regions of the device tree
> > >> > which describe those parts.
> > >> >
> > >> > A test is provided which builds a tree while tracking where the regions
> > >> > should be, then calls fdt_first/next_region() to make sure that it agrees.
> > >> >
> > >> > Further tests will come as part of fdtgrep.
> > >> >
> > >> > Signed-off-by: Simon Glass <sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
> > >>
> > >> Have I sent this to the right place? Any comments?
> > >
> > > Sorry. I know you've resent this several times, and I've been
> > > procrastinating about it since forever.
> > >
> > > Basically, I'm just not convinced. For all your efforts to explain
> > > the rationale, it just seems like a really ad-hoc set of flags and
> > > conditions that doesn't seem to form a coherent whole.
> >
> > Are you referring to fdtgrep or the new library function? I'd be happy
> > enough getting the library function to start with in if there isn't
> > much need for the grep utility.
>
> The library function primarily.
>
> > > From the lack of other responses, I'm assuming there's not really
> > > anyone else who sees it as a compelling feature either.
> >
> > Is there normally a lot of mailing list traffic for new libfdt
> > features? This is used for verified boot in U-Boot for example.
>
> Hm, that's true I guess.
I'd like to dredge this up again.
I'm currently working on adding device tree support in U-Boot SPL (the
tiny loader that loads U-Boot). U-Boot currently uses device tree for
normal operation, but since SPL can run in very small RAM sizes there
is often not enough space for a full (~40KB) device tree. In fact
often all that is needed is the UART and perhaps SPI/I2C. This is
<1KB.
There seem to be a several options here:
1. Create a separate device tree file for each board, just for SPL,
with just what is needed to boot SPL
2. Use #ifdef and the preprocessor to remove what is not needed, and
run dtc twice with different #defines
3. Use fdtgrep to chop down the device tree file by removing unwanted nodes
I think that option 3 is the best. It avoids copying source code, or
littering it with #ifdef.
If necessary I suppose this tool could just live in U-Boot but that
seems a shame. I understand that fdtgrep perhaps has too many options
for your liking. Is there some subset that you think would be worthy
of including as a new device-tree-compiler tool?
If so, then I would rather create fdtgrep (or fdtprune or whatever)
there than add a special tool to U-Boot.
Please let me know what you think.
Regards,
Simon
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2015-05-07 1:08 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-03-22 15:07 [RESEND PATCH v3 0/3] Introduce fdtgrep for subsetting and hashing FDTs Simon Glass
[not found] ` <1395500836-12735-1-git-send-email-sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
2014-03-22 15:07 ` [RESEND PATCH v3 1/3] libfdt: Add function to find regions in an FDT Simon Glass
[not found] ` <1395500836-12735-2-git-send-email-sjg-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
2014-05-01 16:13 ` Simon Glass
[not found] ` <CAPnjgZ0ehgSd1WprpMd1ZY=LaZaKkBG3MaNPV5ja27Q2p12WvA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-05-02 5:40 ` David Gibson
[not found] ` <20140502054014.GF6035-1s0os16eZneny3qCrzbmXA@public.gmane.org>
2014-06-03 3:12 ` Simon Glass
[not found] ` <CAPnjgZ2OHco3GjH_rPC+7Fu36QqBK2_sWvcaF1WOWFpBNkZTcQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-06-05 14:12 ` David Gibson
[not found] ` <20140605141236.GA25412-RXTfZT5YzpxwFLYp8hBm2A@public.gmane.org>
2015-05-07 1:08 ` Simon Glass
2014-03-22 15:07 ` [RESEND PATCH v3 2/3] Add documentation for fdtget/put Simon Glass
2014-03-22 15:07 ` [RESEND PATCH v3 3/3] Add fdtgrep to grep and subset FDTs Simon Glass
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.