In rework:
- size has been removed from array, as it has been making array
at least of two elements, even if scratch partition not used;
- some variables have been localized and some removed;
- no longer returning rc codes from flash_area_ functions on
failure, just single point of failure-exit - it made no sense
anyway as we can not recover from flash failure;
- flash area pointer is now NULLified on failure, or when
status not found.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The commit moves static common functions that have duplicate
definitions in bootutil_public.c and bootutil_misc.c to header
file to be included by both files.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
So far the FIH CI job was fetching a WIP change from
TF-M Gerrit to align the MCUboot and TF-M from the FIH
library point of view. This change is replacing to fetch
the final version instead of the WIP change.
Signed-off-by: Tamas Ban <tamas.ban@arm.com>
Fixes some confusing Kconfig values to have values that make
sense for the underlying SMP transport.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
When hooks are enabled then boot_reset_request_hook will be
called to check whether it is allowed to reset a device.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
Provide prototype for a new hook boot_reset_request_hook
which is called when device is requested to reboot.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The relative links do not work on the docs website due to the files not
living in the docs/ directory. This updates to link directly to the
GitHub source so that users are not dropped onto a 404 page.
Signed-off-by: hasheddan <georgedanielmangum@gmail.com>
When MCUBOOT_SINGLE_APPLICATION_SLOT is set then the app can
only be overwritten with new image and scratch algorithm
is not used. The configuration logic would default
MCUBOOT_SWAP_USING_SCRATCH to 1 because it was lacking
check for MCUBOOT_SINGLE_APPLICATION_SLOT.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
Add support for RT595 to MCUBoot. A larger number of max sectors is
required due to the large flash size present on the RT595 EVK.
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
This deprecates the flash erase Kconfig for zephyr, if this action
is required then the board should be flashed using west with the
`--erase` argument supplied instead.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
Fix a swap corruption which occurs on the swap move algorithm when a
reset happens exactly at the point after the last move up, and its
status update. On restart the image headers should be read at the 2nd
sector of the primary slot, but due to lacking initialization it is
read on the first sector, and then fails. This error was masked on the
simulator because of the use of a global variable, which retained its
value on a "reset simulation".
Fixes#1588
Signed-off-by: Fabio Utzig <utzig@apache.org>
Add jobs for testing build ESP32-XX within more features:
Serial Recovery, Multi Image and Multi Boot
Signed-off-by: Almir Okato <almir.okato@espressif.com>
Returned values are now hardcoded. Indeed, while it is not
strictly needed (few return values different from SUCCESS
or FAILURE) complexity added by encoding return values might
cause the software to be vulnerable to fault attacks.
Return type changed from fih_int to fih_ret to make
the whole thing much simpler and therefore more robust
to fault attacks. In addition, its easier to predict
compiler behavior.
Affectation of sentive variables has been hardened using macro
FIH_SET (affectation + check wether write access has been properly
done). FIH_DECLARE() is added to ease the declaration of sentive
variables.
Equality tests fih_eq() and fih_not_eq() are now macros because
inlining produce more complex code (and weaker) than macros.
In addition fih_not_eq is modified to be the negation of fih_eq
which was not the case until now.
when FIH_NOT_EQ is used , FIH_SET(fih_rc, FIH_FAILURE) has been added
in some part of the code.
variable image_mask (bootutil_priv.h) is now volatile because a
double IF test is made on it.
some others parts of the code have been hardenned (eg. loop on images)
Signed-off-by: Michael Grand <m.grand@trustngo.tech>
This makes it possible to enable timeout (BOOT_SERIAL_WAIT_FOR_DFU) mode
for the serial recovery when using CDC ACM based serial device. This was
runtime tested on nRF52840-Dongle.
Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
A previous change was added to allow the `getpriv` command to dump ec256
keys in both openssl and pkcs8. That PR did not touch other key file
types which resulted in errors using that command with RSA, X25519, etc.
This commit generalizes the passing of the `format` parameter, so each
key type can decide which format it allows a dump to be produced in,
and what default to use.
Fixes#1529
Signed-off-by: Fabio Utzig <utzig@apache.org>
Add basic flash_sector struct and offset calculation routines. This
fixes the build using swap move, because this data is required to
calculate the maximum image size.
Fixes#1567
Signed-off-by: Fabio Utzig <utzig@apache.org>
packaging >= 22 dropped support for LegacyVersion, which was the usual
result of an invalid version number being parsed. Now it is PEP-440
strict and throws an exception on fails. This fixes the script to work
with both older and newer releases.
Signed-off-by: Fabio Utzig <utzig@apache.org>
Fixes a bug when writing to devices which have memory alignment
requirements with data being using directly from a zcbor-response
whereby the alignment of the buffer data does not meet the
requirements of the flash driver.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
Nordic had changed its license identifier to new
more accurate id: LicenseRef-Nordic-5-Clause.
Old identifiers should be updated.
Signed-off-by: Dominik Kilian <Dominik.Kilian@nordicsemi.no>
There are 3 levels of buffers and only the first one seems to be
configurable, this fixes that issue.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
Fixes an issue whereby rc is a signed variable but is returned as
an unsigned variable in the zcbor functions.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
The return value of bootutil_find_key is used as the key_id in the
bootutil_img_validate function, and negative key_id value used in case
of errors. If MCUBOOT_HW_KEY is set, than the key hash is read by
boot_retrieve_public_key_hash function, but the exceptation is only to
return nonzero on failure, so its error code should not be propagated
up to the caller. Instead, bootutil_find_key should return -1 in case
of a platform error.
Change-Id: I8e2bd12a5cf53787e10ae45c2ab556e8a856692d
Signed-off-by: Mark Horvath <mark.horvath@arm.com>
The current specific setting of devicetree overlay files using `set()`
has a couple of built-in flaws.
It keeps readding the overlay file on each subsequent CMake invocation.
The build command (make/ninja), will automatically invoke CMake if there
are any changes to files used as configure time dependencies.
This can easily be seen by manually re-invoking CMake:
```
# First invocation
$ cmake -DBOARD=nrf52840dk_nrf52840 -DDTC_OVERLAY_FILE=custom.dts ..
Loading Zephyr default modules (Zephyr workspace).
-- Application: /projects/github/ncs/bootloader/mcuboot/boot/zephyr
...
-- Found devicetree overlay: custom.dts
-- Found devicetree overlay: bootloader/mcuboot/boot/zephyr/dts.overlay
# Second invocation
$ cmake -DBOARD=nrf52840dk_nrf52840 -DDTC_OVERLAY_FILE=custom.dts ..
Loading Zephyr default modules (Zephyr workspace).
-- Application: /projects/github/ncs/bootloader/mcuboot/boot/zephyr
...
-- Found devicetree overlay: custom.dts
-- Found devicetree overlay: mcuboot/boot/zephyr/dts.overlay
-- Found devicetree overlay: mcuboot/boot/zephyr/dts.overlay
```
Zephyr has built-in support for application specific overlay config
which gets automatically applied when the overlay file is named:
`app.overlay`.
Therefore rename `dts.overlay` to `app.overlay`.
Ref: https://docs.zephyrproject.org/3.2.0/build/dts/howtos.html \
#set-devicetree-overlays
This change further allows users of mcuboot to place their mcuboot
configuration out-of-tree of the sample by using the
`APPLICATION_CONFIG_DIR` setting.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
`--vector-to-sign` only exports the image payload, or digest, to be
signed externally; it doesn't require any keys to be provided. This
commit moves the code outside a key required block, after the payload
and digest were already calculated from "image + headers + protected
TLVs".
Signed-off-by: Fabio Utzig <utzig@apache.org>
The serial recovery depends on CRC from Zephyr, which it should
have been selecting explicitly.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
Fix the remaining PEP8 warnings that appear on image.py, ecdsa.py
or main.py for imgtool.
Signed-off-by: Antonio de Angelis <Antonio.deAngelis@arm.com>
This patch improves the existing ECDSA key generation feature
in the imgtool by:
- Fix a bug in the 'minimal' representation of PKCS#8 keys where
the resulting ASN.1 DER encoding is not compliant
- Add the option to export ECDSA private keys in SEC1 format by
providing a command line option -f or --format that can be
'openssl' (for SEC1 format) or 'pkcs8'. This format ends up in
key encodings which are generally smaller than PKCS#8.
Signed-off-by: Antonio de Angelis <Antonio.deAngelis@arm.com>
Common prints such as the one that happens when an image is signed
are clearer if the source of print is added (i.e. filename), especially
for those build integration flows where imgtool is invoked as part of
a set of operations.
Signed-off-by: Antonio de Angelis <Antonio.deAngelis@arm.com>
Removed the board configuration for Thingy:53 Application Core as it
contains references to the Kconfig modules that are not available in
the upstream Zephyr. The current configuration is set up to work
in the nRF Connect SDK environment and should be moved there.
Signed-off-by: Kamil Piszczek <Kamil.Piszczek@nordicsemi.no>
With the exception of nrf targets BOOT_SWAP_USING_SCRATCH mode was
still the default algorithm.
Changing the preferred mode in cases where no scratch_partition is
defined will allow successfully building mcuboot for such boards w/o
the need for any board specific overlays.
Signed-off-by: Thomas Stranger <thomas.stranger@outlook.com>
Dominik Ermel