Documentation:
* mkeficapsule man-page
UEFI changes:
* add support for signing images to mkeficapsule
* add support for user define capsule GUID
* adjust unit tests for capsules
* fix UEFI image signature validation in case of multiple signatures
-----BEGIN PGP SIGNATURE-----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=AAdJ
-----END PGP SIGNATURE-----
Merge tag 'efi-2022-04-rc2-4' of https://source.denx.de/u-boot/custodians/u-boot-efi
Pull request for efi-2022-04-rc2-4
Documentation:
* mkeficapsule man-page
UEFI changes:
* add support for signing images to mkeficapsule
* add support for user define capsule GUID
* adjust unit tests for capsules
* fix UEFI image signature validation in case of multiple signatures
The previous patch is changing U-Boot's behavior wrt certificate based
binary authentication. Specifically an image who's digest of a
certificate is found in dbx is now rejected. Fix the test accordingly
and add another one testing signatures in reverse order
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
The EFI spec allows for images to carry multiple signatures. Currently
we don't adhere to the verification process for such images.
The spec says:
"Multiple signatures are allowed to exist in the binary's certificate
table (as per PE/COFF Section "Attribute Certificate Table"). Only one
hash or signature is required to be present in db in order to pass
validation, so long as neither the SHA-256 hash of the binary nor any
present signature is reflected in dbx."
With our current implementation signing the image with two certificates
and inserting both of them in db and one of them dbx doesn't always reject
the image. The rejection depends on the order that the image was signed
and the order the certificates are read (and checked) in db.
While at it move the sha256 hash verification outside the signature
checking loop, since it only needs to run once per image and get simplify
the logic for authenticating an unsigned imahe using sha256 hashes.
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Before the capsule authentication is supported, this test script works
correctly, but with the feature enabled, most tests will fail due to
unsigned capsules.
So check the results depending on CAPSULE_AUTHENTICATE or not.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
This test scenario tests a new feature of mkeficapsule, "--guid" option,
which allows us to specify FMP driver's guid explicitly at the command
line.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Since the syntax of mkeficapsule was changed in the previous commit,
we need to modify command line arguments in a pytest script.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
The existing options, "--fit" and "--raw," are only used to put a proper
GUID in a capsule header, where GUID identifies a particular FMP (Firmware
Management Protocol) driver which then would handle the firmware binary in
a capsule. In fact, mkeficapsule does the exact same job in creating
a capsule file whatever the firmware binary type is.
To prepare for the future extension, the command syntax will be a bit
modified to allow users to specify arbitrary GUID for their own FMP driver.
OLD:
[--fit <image> | --raw <image>] <capsule file>
NEW:
[--fit | --raw | --guid <guid-string>] <image> <capsule file>
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add a couple of test cases against capsule image authentication
for capsule-on-disk, where only a signed capsule file with the verified
signature will be applied to the system.
Due to the difficulty of embedding a public key (esl file) in U-Boot
binary during pytest setup time, all the keys/certificates are pre-created.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Now we can use mkeficapsule command instead of EDK-II's script
to create a signed capsule file. So update the instruction for
capsule authentication.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Add a man page for mkeficapsule command.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
With this enhancement, mkeficapsule will be able to sign a capsule
file when it is created. A signature added will be used later
in the verification at FMP's SetImage() call.
To do that, we need specify additional command parameters:
-monotonic-cout <count> : monotonic count
-private-key <private key file> : private key file
-certificate <certificate file> : certificate file
Only when all of those parameters are given, a signature will be added
to a capsule file.
Users are expected to maintain and increment the monotonic count at
every time of the update for each firmware image.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Add CONFIG_TOOLS_MKEFICAPSULE. Then we want to always build mkeficapsule
if tools-only_defconfig is used.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
We need to install libgnutls-devel package to build the host tool,
mkeficapsule, and as of now, there seems to be a depencency conflict
in the current msys2 installer;
:: installing libp11-kit (0.24.1-1) breaks dependency \
'libp11-kit=0.23.22' required by p11-kit
To resolve this conflict, however, the initial "pacman -Syyuu" in
'tools_only_windows' job is not enough. Another "pacman -Su" will
enforce all the out-of-date packages being upgraded.
(Probably the first "-Syyuu" can be changed to "-Syu".)
See the installation steps in
https://www.msys2.org/
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
A partial list:
- fw_env updates, a new testcase for mkimage -o ..., nop-phy reset-gpios
support, DFU updates, kaslr-seed support in extlinux.conf, modern
"partitions" support in mtd device tree
The output size for snprint() should not only be respected for whole fields
but also with fields. Add more tests.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Some usb-nop-xceiv devices use a gpio take them out
of reset. Add a reset function to put them into that
state. This is similar to how Linux handles the
usb-nop-xceiv driver.
Signed-off-by: Adam Ford <aford173@gmail.com>
Environment variables can be stored in two formats:
1. Single entry with header containing CRC32
2. Two entries with extra flags field in each entry header
For that reason fw_env_open() has two main code paths and there are
pointers for CRC32/flags/data.
Previous implementation was a bit hard to follow:
1. It was checking for used format twice (in reversed order each time)
2. It was setting "environment" global struct fields to some temporary
values that required extra comments explaining it
This change simplifies that code:
1. It introduces two clear code paths
2. It sets "environment" global struct fields values only once it really
knows them
To be fair there are *two* crc32() calls now and an extra pointer
variable but that should be cheap enough and worth it.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
It's usually easier to understand code & follow it if all arguments are
passed explicitly. Many coding styles also discourage using global
variables.
Behaviour of flash_io() was a bit unintuitive as it was writing to a
buffer referenced in a global struct. That required developers to
remember how it works and sometimes required hacking "environment"
global struct variable to read data into a proper buffer.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Since dfu is not only used for USB, and some platform only
supports DFU_OVER_TFTP or EFI capsule update, dfu_alt_info
is defined on such platforms too.
For such platform, 'dfu list' command is useful to check
how the current dfu_alt_info setting is parsed.
Signed-off-by: Masami Hiramatsu <masami.hiramatsu@linaro.org>
Fix some typo and wrong information about dfu_alt_info.
- Add the parameter format, decimal only or hexadecimal.
- Use same parameter name for the same kind of parameters.
(e.g. dev -> dev_id)
Signed-off-by: Masami Hiramatsu <masami.hiramatsu@linaro.org>
When parsing the dfu_alt_info, check the number of arguments
and argument string strictly. If there is any garbage data
(which is not able to be parsed correctly) in dfu_alt_info,
that means something wrong and user may make a typo or mis-
understanding about the syntax. Since the dfu_alt_info is
used for updating the firmware, this mistake may lead to
brick the hardware.
Thus it should be checked strictly for making sure there
is no mistake.
Signed-off-by: Masami Hiramatsu <masami.hiramatsu@linaro.org>
If dfu_alt_info has repeated spaces or tab (for indentation or
readability), the dfu fails to parse it. For example, if
dfu_alt_info="mtd nor1=image raw 100000 200000" (double spaces
after "raw"), the image entity start address is '0' and the size
'0x100000'. This is because the repeated space is not skipped.
Use space and tab as a separater and apply skip_spaces() to
skip redundant spaces and tabs.
Signed-off-by: Masami Hiramatsu <masami.hiramatsu@linaro.org>
Use strlcpy() instead of strcpy() to prevent copying the
entity name over the name buffer size.
Signed-off-by: Masami Hiramatsu <masami.hiramatsu@linaro.org>
Addresses the feedback provided on 5902a397d0 ("mkimage: Allow to
specify the signature algorithm on the command line") which raced with
the merge.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
echo -n does not give the intended effect when invoked in macOS through
/bin/sh, which is the shell make uses by default, see
"https://stackoverflow.com/questions/11675070/makefile-echo-n-not-working"
for a detailed explanation. In this case, it resulted in "-n" being
written to env.txt and env.in even though they should be empty, which
caused compilation to fail with "Your board uses a text-file
environment, so must not define CONFIG_EXTRA_ENV_SETTINGS".
This patch prevents the error by replacing echo -n's with touch, as they
are used to create empty files in these cases.
strlen() returns size_t. So we should use %zu to print it.
This avoids incorrect output on 32bit systems.
Fixes: 2fc62f2991 ("stackprot: Make our test a bit more complex")
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Stress the '-o algo_name' argument of mkimage by expanding the vboot
test.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
[trini: Update scripts/pylint.base]
Replace %zx by %lx and cast size_t to ulong.
U-Boot currently prints garbage debug output:
size=x, ptr=18, limit=18: 4002a000
With this change it prints correct debug data:
size=18, ptr=18, limit=2000: 4002a000
Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Listing MTD partitions directly in the flash mode has been deprecated
for a while for kernel Device Trees. Look for a node "partitions" in the
found flash nodes and use it instead of the flash node itself for the
partition list when it exists, so Device Trees following the current
best practices can be fixed up.
Signed-off-by: Matthias Schiffer <matthias.schiffer@ew.tq-group.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
In the Linux implementation of adc-keys
(drivers/input/keyboard/adc-keys.c), `press-threshold-microvolt` is not
really interpreted as a threshold, but rather as the "nominal voltage"
of the button. When the voltage read from the ADC is closest to a
button's `press-threshold-microvolt`, the button is considered pressed.
This patch reconciles the behavior of button-adc with Linux's adc-keys
such that device trees can be synchronized with minimal modifications.
Signed-off-by: Peter Cai <peter@typeblog.net>
this will add kaslrseed keyword to sysboot lable,
when it set, it will request to genarate random number
from hwrng as kaslr-seed.
with this patch exlinux.conf label looks like
label l0
menu testing
linux /boot/vmlinuz-5.15.16-arm
initrd /boot/initramfs-5.15.16-arm.img
fdtdir /boot/dtbs/5.15.16-arm/
kaslrseed
append root=UUID=92ae1e50-eeeb-4c5b-8939-7e1cd6cfb059 ro
Tested on Khadas VIM with kernel 5.16.0-rc5-arm64, Debian 11.
Signed-off-by: Zhang Ning <zhangn1985@qq.com>
The ci_req->hw_buf can be NULL, test whether it is and if so,
avoid accessing it. Else, the system may crash.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Peter Chen <peter.chen@nxp.com>
Cc: Li Jun <jun.li@nxp.com>
Cc: Peng Fan <peng.fan@nxp.com>
Using the XHCI driver, the function `usb_kbd_poll_for_event` takes
30-40ms to run. The exact time is dependent on the polling interval the
keyboard requests in its descriptor, and likely cannot be significantly
reduced without major rework to the XHCI driver.
The U-Boot EFI console service sets a timer to poll the keyboard every 5
microseconds, and this timer is checked every time a block is read off
disk. The net effect is that, on my system, loading a ~40MiB kernel and
initrd takes about 62 seconds with a slower keyboard and 53 seconds
with a faster one, with the vast majority of the time spent polling the
keyboard.
To solve this problem, this patch adds a 20ms delay between consecutive
calls to `usb_kbd_poll_for_event`. This is sufficient to reduce the
total loading time to under half a second for both keyboards, and does
not impact the perceived keystroke latency.
Signed-off-by: Thomas Watson <twatson52@icloud.com>
The imx8mm and imx8mn appear compatible with imx7d-usb
flags in the OTG driver. If the dr_mode is defined as
host or peripheral, the device appears to operate correctly,
however the auto host/peripheral detection results in an error.
The solution isn't just adding checks for imx8mm and imx8mn to
the check for imx7, because the USB clock needs to be running
to read from the USBNC_PHY_STATUS_OFFSET register or it will hang.
Marek requested that I not enable the clocks in ehci_usb_of_to_plat,
so I modified that function to return an unknown state if the
device tree does not explicitly state whether it is a host
or a peripheral.
When the driver probes, it looks to see if it's in the unknown
state, and only then will it read the register to auto-detect.
Signed-off-by: Adam Ford <aford173@gmail.com>
Tested-by: Tim Harvey <tharvey@gateworks.com>
EMIF tool for AM64 SK is now updated to 0.8.0 that includes
* disabled Write DQ training
* improve CA ODT to 60 ohms
The lpddr4 enabled with periodic WDQ training is causing periodic 26us
stall. This makes the SoC stall without doing anything which leads to
R5 interrupt latency in TCM memory. Due to this periodic training there
are some outstanding CPU transactions waiting for the lpddr4 to complete.
Hence, disable the periodic write DQ training during the
non-initialization stage of lpddr4 which results in an approximate 1us
stall. Also, update the lpddr4 config to improve CA ODT by 60 ohms
The rationales are as follows:
- PI_WDQLVL_EN: 2 Bits register field to support write DQ leveling,
disable bit 1 that supports Write DQ during non-initialization to
avoid ~26us stall during code execution.
- MR11_DATA_F1/F2_x register fields value changed to 0x66 that changes
the CA ODT from 48ohm to 60ohm to improve the eye margin on CA bus by
increasing the signal swing.
Signed-off-by: James Doublesin <doublesin@ti.com>
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
There are no boards that define CONFIG_SYS_RESET_ADDR, so drop the
remaining comments referencing it and also the config_whitelist.txt entry.
Signed-off-by: Ovidiu Panait <ovidiu.panait@windriver.com>
Reviewed-by: Stefan Roese <sr@denx.de>
Acked-by: thomas@wytron.com.tw
There are no boards that define CONFIG_SYS_RESET_ADDRESS, so drop the
associated mpc8xx code that checks for it.
Signed-off-by: Ovidiu Panait <ovidiu.panait@windriver.com>
Reviewed-by: Stefan Roese <sr@denx.de>
Acked-by: Christophe Leroy <christophe.leroy@csgroup.eu>
This driver adds support for the keyboard on Apple Silicon laptops.
The controller for this keyboard sits on an SPI bus and uses an
Apple-specific HID over SPI protocol. The packets sent by this
controller for key presses and key releases are fairly simple and
are decoded directly by the code in this driver and converted into
standard Linux keycodes. The same controller handles the touchpad
found on these laptops. Packets for touchpad events are simply
ignored.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested on: Macbook Air M1
Tested-by: Simon Glass <sjg@chromium.org>
Add a driver for the SPI controller integrated on Apple SoCs.
This is necessary to support the keyboard on Apple Silicon laopts
since their keyboard uses an Apple-specific HID over SPI protocol.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested on: Macbook Air M1
Tested-by: Simon Glass <sjg@chromium.org>
Add a driver for the NVMe storage controller integrated on
Apple SoCs. This NVMe controller isn't PCI based and deviates
from the NVMe standard in its implementation of the command
submission queue and the integration of an NVMMU that needs
to be managed. This commit tweaks the core NVMe code to
support the linear command submission queue implemented by
this controller. But setting up the submission queue and
managing the NVMMU controller is handled by implementing
the driver ops that were added in an earlier commit.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Tested-on: firefly-rk3399
Tested-by: Mark Kettenis <kettenis@openbsd.org>
Tested on: Macbook Air M1
Tested-by: Simon Glass <sjg@chromium.org>
The power management controller found on Apple SoCs als provides
a way to reset all devices within a power domain. This is needed
to cleanly shutdown the NVMe controller before we hand over
control to the OS.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested on: Macbook Air M1
Tested-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
Add a function to disable the NVMe controller. This will be used
to let the driver for the NVMe storage integrated on Apple SoCs
shutdown the NVMe controller such we can shutdown the NVMe
IOP controller in a clean way afterwards before handing control
to the OS.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested on: Macbook Air M1
Tested-by: Simon Glass <sjg@chromium.org>
The NVMe storage controller integrated on Apple SoCs deviates
from the NVMe standard in two aspects. It uses a "linear"
submission queue and it integrates an NVMMU that needs to be
programmed for each NVMe command. Introduce driver ops such
that we can set up the linear submission queue and program the
NVMMU in the driver for this strange beast.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested on: Macbook Air M1
Tested-by: Simon Glass <sjg@chromium.org>
Most Apple IOPs run a firmware that is based on what Apple calls
RTKit. RTKit implements a common mailbox protocol. This code
provides an implementation of the AP side of this protocol,
providing a function to initialize RTKit-based firmwares as well
as a function to do a clean shutdown of this firmware.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested on: Macbook Air M1
Tested-by: Simon Glass <sjg@chromium.org>
U-Boot is expected to support multiple generations of Apple SoCs
in a single binary with a single defconfig. Therefore it makes
more sense to set SYS_SOC to "apple".
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
