Add new feature that allows testing EC256 encrypted images using the
Mbed TLS backend.
Move config-ecdsa.h to config-ec.h because definitions are very similar
between ECDSA and ECDH with Mbed TLS so resort to a single config file.
Add new feature and fix the build; add proper Mbed TLS memory
initialization when enc-ec256-mbedtls is used.
Signed-off-by: Fabio Utzig <utzig@apache.org>
Add Mbed TLS ECDSA signature verification as an option (in addition to
Tinycrypt and the CC310 hardware version). Although the Mbed TLS ECDSA
verification code is both larger and slower, this will still save space
if there is another reason that the Mbed TLS code is already being
brought in for another reason (such as certificate management, for
example).
Mbed TLS's ECDSA verification works at a different level than the other
two libraries, so this takes a bit of reworking. There are some
additional parameters passed to the various functions, and a new define
MCUBOOT_ECDSA_NEED_ASN1_SIG to indicate that the ecdsa verification
wants the original ASN1 signature, not a decoded key.
This adds the boot changes and simulator support to test this configuration.
Signed-off-by: David Brown <david.brown@linaro.org>
Upgrade to the latest aes-ctr package, and apply minor fixes due to
trait naming changes in this version.
Signed-off-by: David Brown <david.brown@linaro.org>
Update to the newest version of this crate. There are no API changes
affecting us, so this is just an update of the Cargo.toml file.
Signed-off-by: David Brown <david.brown@linaro.org>
Updates a few sim dependencies, reducing the amount of crates
required for a build; also gets rid of cases of more than one
version required for some crates. Results in depending on 76
crates instead of 89 before the PR.
Signed-off-by: Fabio Utzig <utzig@apache.org>
To start adding support for devices with a large write size, modify the
simulator to allow a 'large-write' feature that will also run tests with
128 and 512 byte write alignments.
This requires fixing some places in image generation that assumed a
fairly small alignment (namely, padding the buffer at the end, rather
than individual parts), and detecting the large writes when writing the
MAGIC value to the image.
The flash API still passes the alignment back as a uint_8, but in
overwrite mode, this verifies that the boot code won't use this value.
Currently the 'large-write' feature requires 'overwrite-only' as well.
Signed-off-by: David Brown <david.brown@linaro.org>
Instead of manually expanding multi-byte encoding, use the byteorder
crate which has its own extension methods to do this. This both makes
the code a bit clearer, and also makes it clear that these encodings are
specific to little endian platforms.
Signed-off-by: David Brown <david.brown@linaro.org>
This change replaces the slot 0/1 terminology with primary/secondary
slot and replaces FLASH_AREA_IMAGE_0/1 with
FLASH_AREA_IMAGE_PRIMARY/SECONDARY. This naming convention may be more
understandable, fits better to MCUs with multiple images and it is an
architecture agnostic alternative as well.
Change-Id: I655a585f6ae023852c671ee6635399efe25209c9
Signed-off-by: David Vincze <david.vincze@arm.com>
Signed-off-by: David Brown <david.brown@linaro.org>
This removes the unsafe Tinycrypt bindings previously used for signing
with ECDSA, and relies on ring native support.
The ring library was updated to 0.14.1.
Signed-off-by: Fabio Utzig <utzig@apache.org>
Since logging was broken due to incompatibility between log and
env_logger versions, those crates were updated to known to be compatible
versions. Update initialization of env_logger that does not return a
Result<> anymore.
Other crates were updated to remove duplicated versions as much as
possible.
Signed-off-by: Fabio Utzig <utzig@apache.org>
Automatic migration to Rust 2018. This is the result of running
cargo fix --edition
The resulting code is compatible with both Rust 2015 and 2018. Change
the edition field in the Cargo.toml file as well to begin a more
complete migration.
Signed-off-by: David Brown <david.brown@linaro.org>
A few packages have updates that make them more convient to use with
Rust 2018's 'use macro'. This is convenient as it allows control over
what macros are imported, but without these updates, these packages
require you to know the names of internal macros to import.
Signed-off-by: David Brown <david.brown@linaro.org>
This adds new cargo features to allow running tests of encrypted
images with both RSA-OAEP and AES-128-KW.
When installing images on the simulated flash, both a plain and an
encrypted images are created. When encrypted image support is enabled,
verification of images in slot1 match against the encrypted image,
otherwise plain images are used.
PS: Also fixes ImageHeader to match bootutil definition.
Signed-off-by: Fabio Utzig <utzig@apache.org>
This updates error-chain to remove a warning on recent rust distros.
`pem` was update to get fairly recent `error-chain` as well.
Signed-off-by: Fabio Utzig <utzig@apache.org>
As a start of doing the testing using Rust/Cargo's test framework, write
a test runner that just runs the existing tests. They run as they do
now, except that there is an assertion that there were no failures.
Signed-off-by: David Brown <david.brown@linaro.org>
Adds the feature to enable testing of overwrite-only functionality.
While running with this feature enabled, disable tests that try to
revert and disables trailer verification, since overwrite-only doesn't
rely on status writing.
Signed-off-by: Fabio Utzig <utzig@apache.org>
Remove the key_id field from the image header. There are two problems
with this field. First, it is only an integer offset, and so causes an
unnecessarily tight coupling between the particular keys built into the
bootloader, and the key that is used to sign. Second, it makes the
key_id part of the image header, which is included in the signature.
This makes it impossible to later sign the image with a different
signature.
Instead of the key-id, add a TLV KEYHASH entry. This will hold the
SHA256 of the public key that the signature is against. Each signature
placed in the TLV should be preceeded by this entry to indicate the
public key used.
The signature check will check each signature, and if the KEYHASH is
known and the signature type is supported, it will be checked. As long
as at least one signature is considered valid, the image will be
considered signed. This also allows the image to be signed with
multiple signatures to support having different devices with possibly
different keys compiled into the bootloaders.
Based on work by Marko Kiiskila <marko@runtime.io>
Signed-off-by: Marko Kiiskila <marko@runtime.io>
Signed-off-by: David Brown <david.brown@linaro.org>
The docopt package has moved from rustc-serialize to serde. Move to
this version, upgrading the custom deserialize we use for the alignment
argument.
Signed-off-by: David Brown <david.brown@linaro.org>
Move most dependencies to the latest version at the time of this commit.
docopt is left at the current version as there is an api change in the
newer version.
Signed-off-by: David Brown <david.brown@linaro.org>
Create a crate `mcuboot-sys` to hold the building and binding of the
mcuboot code. There aren't any substantive code changes here, just
moving the code into a separate crate.
Signed-off-by: David Brown <david.brown@linaro.org>
Using the published debug keypair, generate a signature with it, to add
to the TLV. This verifies that signature verification works as
expected.
Signed-off-by: David Brown <david.brown@linaro.org>
The bitflags crate seems to be better supported, and doesn't have the
problems getting the results back into the enum type.
Signed-off-by: David Brown <david.brown@linaro.org>
The ring crate's SHA256 is slightly slower than the one from
rust-crypto, but is much cleaner to use. It also has all of the hashing
operations we need.
This crate is somewhat opinionated, e.g. will not sign messages without
also performing the signature itself. Unfortunately, this means we'll
end up computing signatures of the same data many times, and perhaps
this should be improved, since we're always using the same data.
Signed-off-by: David Brown <david.brown@linaro.org>
Instead of stubbing out the image validation code, compile it, and add
the SHA256 TLV to the buffer.
Signed-off-by: David Brown <david.brown@linaro.org>
In preparation for moving the bootutil binding into a separate crate,
move the flash simulator into its own crate. This will allow the
binding to have access to the simulator without creating a circular
dependency.
Signed-off-by: David Brown <david.brown@linaro.org>
Use the Rust logging so that normal messaging isn't overwhelmed by
messages printed. The default level is 'error' which will only print
full errors (including the failures at the end).
To run with warnings:
RUST_LOG=warn ./target/release/bootsim ...
This will print warnings, showing the particular failures in mind.
Using 'info' instead of 'warn' will print lots of progress as it runs.
'sim' is a small simulator for the bootloader's update code. It tests
untimely powerdowns to ensure that the bootloader will recover from a
power loss or reset at any time during the boot.
Note that, as of this commit, there are some failures in the test that
need to be investigated.
Also note that this build script does not output proper dependencies for
source files outside of the simulator directory, and won't rebuild the C
files if they or headers are modified.
Fabio Utzig