The hash algorithm selection was streamlined in commit 92055e138f
("image: Drop if/elseif hash selection in calculate_hash()"). Said
commit kept the call to cpu_to_uimage() to convert the CRC to big
endian format.
This would have been correct when calling crc32_wd(). However, the
->hash_func_ws member of crc32 points to crc32_wd_buf(), which already
converts the CRC to big endian. On a little endian host, doing both
conversions results in a little-endian CRC. This is incorrect.
To remedy this, simply drop the call to cpu_to_uimage(), thus only
doing the byte-order conversion once.
Fixes: 92055e138f ("image: Drop if/elseif hash selection in
calculate_hash()")
Tested-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
calculate_hash() would try to select the appropriate hashing function
by a if/elseif contruct. But that is exactly why hash_lookup_algo()
exists, so use it instead.
This does mean that we now have to 'select HASH' to make sure we get
the hash_lookup_algo() symbol. However, the change makes sense because
even basic FITs will have to deal with "hash" nodes.
My only concern is that the 'select SPL_HASH' might cause some
platform to grow above its SPL size allowance
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
[trini: Make FSL_CAAM be implied only on ARM && SPL]
Signed-off-by: Tom Rini <trini@konsulko.com>
It is a pain to have to specify the value 16 in each call. Add a new
hextoul() function and update the code to use it.
Add a proper comment to simple_strtoul() while we are here.
Signed-off-by: Simon Glass <sjg@chromium.org>
Switch to fit_image_get_data_and_size() for consistency with all other
data loaded from FIT.
Signed-off-by: John Keeping <john@metanate.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
If signature verification is not in use we don't need to worry about
the risk of using @ in node names. Update fit_image_verify() to allow
it if the function is not enabled.
Signed-off-by: Simon Glass <sjg@chromium.org>
This is not needed with Kconfig, since we can use IS_ENABLED() easily
enough. Drop it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
We already have a host Kconfig for these SHA options. Use
CONFIG_IS_ENABLED(SHAxxx) directly in the code shared with the host build,
so we can drop the unnecessary indirections.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
We already have a host Kconfig for SHA1. Use CONFIG_IS_ENABLED(SHA1)
directly in the code shared with the host build, so we can drop the
unnecessary indirection.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
The current fitImage DTO implementation expects each fitImage image
subnode containing DTO to have 'load' property, pointing somewhere
into memory where the DTO will be loaded. The address in the 'load'
property must be different then the base DT load address and there
must be sufficient amount of space between those two addresses.
Selecting and using such hard-coded addresses is fragile, error
prone and difficult to port even across devices with the same SoC
and different DRAM sizes.
The DTO cannot be applied in-place because fdt_overlay_apply_verbose()
modifies the DTO when applying it onto the base DT, so if the DTO was
used in place within the fitImage, call to fdt_overlay_apply_verbose()
would corrupt the fitImage.
Instead of copying the DTO to a specific hard-coded load address,
allocate a buffer, copy the DTO into that buffer, apply the DTO onto
the base DT, and free the buffer.
The upside of this approach is that it is no longer necessary to
select and hard-code specific DTO load address into the DTO. The
slight downside is the new malloc()/free() overhead for each DTO,
but that is negligible (*).
(*) on iMX8MM/MN and STM32MP1
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Pantelis Antoniou <pantelis.antoniou@konsulko.com>
Cc: Simon Glass <sjg@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
[trini: Add <linux/sizes.h>]
Signed-off-by: Tom Rini <trini@konsulko.com>
board_fit_image_post_process() passes only start and size of the image,
but type of the image is not passed. So pass fit and node_offset, to
derive information about image to be processed.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Tero Kristo <kristo@kernel.org>
OP-TEE images are normally packaged with
type = "tee;
os = "tee";
However, fit_image_load() thinks that is somehow invalid. However if
they were declared as type = "kernel", os = "linux", fit_image_load()
would happily accept them and allow the boot to continue. There is no
technical limitation to excluding "tee".
Allowing "tee" images is useful in a boot flow where OP-TEE is
executed before linux.
In fact, I think it's unintuitive for a "load"ing function to also do
parsing and contain a bunch ad-hoc heuristics that only its caller
might know. But I don't make the rules, I just write fixes. In more
polite terms: refactoring the fit_image API is beyond the scope of
this change.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Consider the following FIT:
images {
whipple {};
};
configurations {
conf-1 {
firmware = "whipple";
};
};
Getting the 'firmware' image with fit_image_load() is not possible, as
it doesn't understand 'firmware =' properties. Although one could pass
IH_TYPE_FIRMWARE for 'image_type', this needs to be converted to a
"firmware" string for FDT lookup -- exactly what this change does.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Sandbox is special in that it is used for testing and it does not match
any particular target architecture. Allow it to load an image from any
architecture, so that 'bootm' can be used as needed.
Signed-off-by: Simon Glass <sjg@chromium.org>
Unfortunately -ENODATA is not available in OpenBSD. Use -EBADMSG
instead, to indicate a missing timestamp.
Fixes: c5819701a3 image: Adjust the workings of fit_check_format()
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
Using unit addresses in a FIT is a security risk. Add a check for this
and disallow it.
CVE-2021-27138
Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Bruce Monroe <bruce.monroe@intel.com>
Reported-by: Arie Haenel <arie.haenel@intel.com>
Reported-by: Julien Lenoir <julien.lenoir@intel.com>
Some strange modifications of the FIT can introduce security risks. Add an
option to check it thoroughly, using libfdt's fdt_check_full() function.
Enable this by default if signature verification is enabled.
CVE-2021-27097
Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Bruce Monroe <bruce.monroe@intel.com>
Reported-by: Arie Haenel <arie.haenel@intel.com>
Reported-by: Julien Lenoir <julien.lenoir@intel.com>
At present this function does not accept a size for the FIT. This means
that it must be read from the FIT itself, introducing potential security
risk. Update the function to include a size parameter, which can be
invalid, in which case fit_check_format() calculates it.
For now no callers pass the size, but this can be updated later.
Also adjust the return value to an error code so that all the different
types of problems can be distinguished by the user.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Bruce Monroe <bruce.monroe@intel.com>
Reported-by: Arie Haenel <arie.haenel@intel.com>
Reported-by: Julien Lenoir <julien.lenoir@intel.com>
When searching for a node called 'fred', any unit address appended to the
name is ignored by libfdt, meaning that 'fred' can match 'fred@1'. This
means that we cannot be sure that the node originally intended is the one
that is used.
Disallow use of nodes with unit addresses.
Update the forge test also, since it uses @ addresses.
CVE-2021-27138
Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Bruce Monroe <bruce.monroe@intel.com>
Reported-by: Arie Haenel <arie.haenel@intel.com>
Reported-by: Julien Lenoir <julien.lenoir@intel.com>
Move this out of the common header and include it only where needed. In
a number of cases this requires adding "struct udevice;" to avoid adding
another large header or in other cases replacing / adding missing header
files that had been pulled in, very indirectly. Finally, we have a few
cases where we did not need to include <asm/global_data.h> at all, so
remove that include.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Config allows to disable printing contents of fitImage to optimize boottime.
Signed-off-by: Ravik Hasija <rahasij@linux.microsoft.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
When CONFIG_FIT_CIPHER=y and CONFIG_FIT_SIGNATURE=n is there is no
implementation of image_get_host_blob for mkimage/dumpimage:
/usr/bin/ld: tools/common/image-cipher.o: in function `fit_image_decrypt_data':
image-cipher.c:(.text+0x9a): undefined reference to `image_get_host_blob'
Move the implementation to a common file so it can be shaed between
image-cipher.c and image-fit-sig.c.
Signed-off-by: Joel Stanley <joel@jms.id.au>
fit_check_format() must check that the buffer contains a flattened device
tree before calling any device tree library functions.
Failure to do may cause segmentation faults.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Support reusing board_fit_config_name_match() to automatically
select a sensible default configuration for booting fitImages
using 'bootm'.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Replace most #ifdef checks for USE_HOSTCC and CONFIG_*
with normal if instructions.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
The commit 9f45aeb937 ("spl: fit: implement fdt_record_loadable") which
introduced fdt_record_loadable() state there spl_fit.c is not 64bit safe.
Based on my tests on Xilinx ZynqMP zcu102 platform there shouldn't be a
problem to record these addresses in 64bit format.
The patch adds support for systems which need to load images above 4GB.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The current recommendation for best security practice from the US government
is to use SHA384 for TOP SECRET [1].
This patch adds support for SHA384 and SHA512 in the hash command, and also
allows FIT images to be hashed with these algorithms, and signed with
sha384,rsaXXXX and sha512,rsaXXXX
The SHA implementation is adapted from the linux kernel implementation.
[1] Commercial National Security Algorithm Suite
http://www.iad.gov/iad/programs/iad-initiatives/cnsa-suite.cfm
Signed-off-by: Reuben Dowle <reuben.dowle@4rf.com>
These are used in multiple places so update them to use a shared #define.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philippe Reynes <philippe.reynes@softathome.com>
It is useful to be a little more specific about what is being checked.
Update a few messages to help with this.
Signed-off-by: Simon Glass <sjg@chromium.org>
This adds the check against IH_OS_VXWORKS during FIT image load,
to allow loading FIT image for VxWorks.
Signed-off-by: Lihua Zhao <lihua.zhao@windriver.com>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
GCC-10 reports:
In file included from tools/common/image-fit.c:1:
include/image.h: In function ‘fit_image_get_data_and_size’:
./tools/../common/image-fit.c:1015:9: warning: ‘len’ may be used
uninitialized in this function [-Wmaybe-uninitialized]
1015 | *size = len;
| ~~~~~~^~~~~
./tools/../common/image-fit.c:996:6: note: ‘len’ was declared here
996 | int len;
| ^~~
Add the missing check of the return value of fit_image_get_data_size().
Fixes: c3c8638804 ("add FIT data-position & data-offset property support")
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Introduce new configuration, CONFIG_RSA_VERIFY which will decouple building
RSA functions from FIT verification and allow for adding a RSA-based
signature verification for other file formats, in particular PE file
for UEFI secure boot.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
This commit add to u-boot the support to decrypt
fit image encrypted with aes. The FIT image contains
the key name and the IV name. Then u-boot look for
the key and IV in his device tree and decrypt images
before moving to the next stage.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
This commit add the support of encrypting image with aes
in mkimage. To enable the ciphering, a node cipher with
a reference to a key and IV (Initialization Vector) must
be added to the its file. Then mkimage add the encrypted
image to the FIT and add the key and IV to the u-boot
device tree.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Add a new OS type to be used for chain-loading an EFI compatible
firmware or boot loader like GRUB2, possibly in a verified boot
scenario.
Bellow is sample ITS file that generates a FIT image supporting
secure boot. Please note the presence of 'os = "efi";' line, which
identifies the currently introduced OS type:
/ {
#address-cells = <1>;
images {
efi-grub {
description = "GRUB EFI";
data = /incbin/("bootarm.efi");
type = "kernel_noload";
arch = "arm";
os = "efi";
compression = "none";
load = <0x0>;
entry = <0x0>;
hash-1 {
algo = "sha256";
};
};
};
configurations {
default = "config-grub";
config-grub {
kernel = "efi-grub";
signature-1 {
algo = "sha256,rsa2048";
sign-images = "kernel";
};
};
};
};
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@gmail.com>
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Drop inclusion of crc.h in common.h and use the correct header directly
instead.
With this we can drop the conflicting definition in fw_env.h and rely on
the crc.h header, which is already included.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
The Linux ramdisk should always be decompressed by the kernel itself,
not by U-Boot. Therefore, the 'compression' node in the FIT image should
always be set to "none" for ramdisk images, since the only point of
using that node is if you want U-Boot to do the decompression itself.
Yet some systems populate the node to the compression algorithm used by
the kernel instead. This used to be ignored, but now that we support
decompression of all image types it becomes a problem. Since ramdisks
should never be decompressed by U-Boot anyway, this patch adds a special
exception for them to avoid these issues. Still, setting the
'compression' node like that is wrong in the first place, so we still
want to print out a warning so that third-party distributions doing this
can notice and fix it.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Tested-by: Heiko Schocher <hs@denx.de>
Reviewed-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
This patch adds support for an optional optimization to compatible
string matching where the compatible string property from the root node
of the kernel FDT can be copied into the configuration node of the FIT
image. This is most useful when using compressed FDTs or when using FDT
overlays, where the traditional extraction of the compatible string from
the kernel FDT itself is not easily possible.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
This patch adds support for compressing non-kernel image nodes in a FIT
image (kernel nodes could already be compressed previously). This can
reduce the size of FIT images and therefore improve boot times
(especially when an image bundles many different kernel FDTs). The
images will automatically be decompressed on load.
This patch does not support extracting compatible strings from
compressed FDTs, so it's not very helpful in conjunction with
CONFIG_FIT_BEST_MATCH yet, but it can already be used in environments
that select the configuration to load explicitly.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Currently only the first fdt is loaded from the extra configuration of
FIT image.
If the configuration have multiple fdt, load them all as well.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
The rsa signature use a padding algorithm. By default, we use the
padding pkcs-1.5. In order to add some new padding algorithm, we
add a padding framework to manage several padding algorithm.
The choice of the padding is done in the file .its.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Rather than verifying configuration signature of the configuration node
containing the kernel image types, verify all configuration nodes, even
those that do not contain kernel images. This is useful when the nodes
contain ie. standalone OSes or U-Boot.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Tom Rini <trini@konsulko.com>
Cc: Pantelis Antoniou <pantelis.antoniou@konsulko.com>
Cc: Simon Glass <sjg@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add FIT data-position & data-offset property support for bootm,
which were already supported in SPL.
Signed-off-by: Kelvin Cheung <keguang.zhang@gmail.com>
Just add IH_TYPE_STANDALONE to fit_get_image_type_property().
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Pantelis Antoniou <pantelis.antoniou@konsulko.com>
Cc: Simon Glass <sjg@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
These functions may be needed in SPL, so add empty variants of them
if CONFIG_SPL_FIT_PRINT is disabled.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Pantelis Antoniou <pantelis.antoniou@konsulko.com>
Cc: Simon Glass <sjg@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
